Seventy-four patients undergoing anatomic or reverse total shoulder arthroplasty were randomized to receive either a non-opioid multimodal pain protocol (n=37) or a multimodal protocol with 28 tablets of 5 mg oxycodone (n=37). The primary outcome of interest was opioid utilization (measured in morphine milligram equivalents). Secondary outcomes of interest included visual analog scale (VAS) pain scores, PROMIS Pain Interference scores, and adverse medication side effects. Outcomes were assessed over 10 days postoperatively. Overall, the results of the study revealed no significant differences in pain control between groups, but a substantial reduction in opioid consumption in the non-opioid group. The prevalence of constipation was also significantly lower in the non-opioid cohort. These findings suggest that non-opioid multimodal protocols can offer effective pain control while minimizing opioid exposure and associated side effects after shoulder arthroplasty.