Background. Nurse-initiated pain protocols (NIPPs) at emergency department (ED) triage remain underused. This study investigated factors associated with patient refusal and nurse use of NIPP, accounting for triage operational context. Methods. This retrospective observational study combined prospectively collected nurse characteristics with retrospective data on NIPP use over 15 months in a tertiary university hospital ED. Outcomes included rates of NIPP refusal and use, documented reasons for refusal, and associations with patient characteristics, nurse characteristics, crowding, and operational pressure. Results. Sixty-three triage nurses managed 16,137 adult patients; 6.2% refused the NIPP. Among consenting patients, NIPP was used in one-third of encounters. Multi-level logistic regression revealed significant variation between nurses in both refusal and use. Refusal was more likely in patients with lower acuity and among nurses trained in Europe or concerned about prescribing responsibility, but less frequent with severe pain or longer triage duration. NIPP use was more frequent with lower acuity, higher pain intensity, longer triage duration, crowding, and among nurses with European training, but decreased in older patients and those arriving by ambulance. Conclusions. NIPP refusal and use at triage were both low, with marked variability between nurses. Patient characteristics and triage operational factors were most strongly associated with outcomes, while nurse-related factors contributed less. These findings support prospective implementation studies to clarify drivers of practice variation and optimize analgesia delivery at triage..