Cannabis legalization has increased product availability and use, subsequently necessitating efficient measurements of cannabis use that accurately reflect biomarkers of cannabis exposure (e.g. blood cannabinoid levels). The present study thus sought to compare cannabis use metrics and their associations with biomarkers and examine whether these associations were moderated by sex or age. Observational study using data from five larger studies. Data collection and recruitment occurred in the greater Boulder/Denver metropolitan area in Colorado, USA. Individuals (n = 1090, Mage = 32.89, SD = 12.97; 78.35% White; 51.56% female) included regular cannabis users (Muse = 16 days/past month and 4 times/day). Participants completed assessments of typical quantity and frequency of cannabis use via an in-house survey (Cannabis Quantity and Frequency Scale; CQFS) versus past month use via the Timeline Follow Back (TLFB). Cannabis biomarkers were also collected, including blood levels of delta-9 tetrahydrocannabinol (THC) after immediate use and baseline levels of the primary THC metabolite, 11-nor-9-carboxy-tetrahydrocannabinol (THC-COOH) FINDINGS: Separate mixed effects models using TLFB versus CQFS cannabis metric predictors of the two biomarkers including moderators of sex and age resulted in higher adjusted R2 values for the CQFS versus TLFB model predicting THC-COOH (0.30 vs 0.27, respectively) and for the TLFB versus CQFS model predicting THC (0.24 vs 0.21, respectively). Additionally, greater CQFS total times of any cannabis/day (i.e. total times/day) was associated with increased THC-COOH B = 5.85, 95% confidence interval (CI) = 0.66-11.03, P = 0.03 while it was associated with increased THC among males only (Binteraction = 7.80, 95% CI = 0.25-15.34, P = 0.04). TLFB days/month was statistically significantly, positively related with both THC-COOH and THC (B = 2.39, 95% CI = 1.00-3.79, P < 0.001 and B = 4.18, 95% CI = 0.98-7.38, P = 0.01, respectively) with no statistically significant interactions. The Cannabis Quantity and Frequency Scale (CQFS) measurement of total times/day appears to be a better predictor of general cannabis use than the Timeline Follow Back (TLFB) method. In contrast, TLFB appears to be better at predicting acute use of cannabis than the CQFS.
Skrzynski et al. (Fri,) studied this question.