Sex Differences in Opioid Addiction

Body weight during the pre-treatment period was analyzed by 2 × 3 × 6 (sex × pre-treatment condition × pre-treatment day) repeated measures ANOVA. Adult weight was analyzed by 2 × 3 (sex × pre-treatment condition) ANOVA. The three baseline paw-lick trials were averaged and analyzed by 2 × 3 (sex × pre-treatment condition) ANOVAs. Data from the postdrug paw-lick assessment were also averaged over the three test trials but were analyzed by Kruskal–Wallis and Mann–Whitney tests because these data did not meet the normality of distribution or homogeneity of variance assumptions. The total nose pokes and days to criterion were analyzed for sucrose training using 2 × 3 (sex × pre-treatment condition) ANOVAs. Total nose pokes and days to criterion for acquisition training phases 1–4 were analyzed using 2 × 3 × 4 (sex × pre-treatment condition × training phase) repeated measures ANOVAs. Total nose pokes for the seven-phase 5 testing days (fentanyl-only sessions) were analyzed using 2 × 3 × 7 (sex × pre-treatment condition × day) repeated measures ANOVAs. Significant higher-order interactions were analyzed using lower-order ANOVAs. Post hoc analysis of data was made using Tukey’s tests (p < 0.05). Effect sizes were reported as partial eta squared (η_p²) and categorized based on the following scale: η_p² ≤ 0.03 (small effect), η_p² > 0.03 and ≤0.10 (medium effect), and η_p² > 0.10 (large effect) (Labots et al., 2016 (link)).