Ability of pain scoring scales to differentiate between patients desiring analgesia and those who do not in the emergency department

Abstract

Background and Importance: Pain is one of the most reasons for a visit to an emergency department (ED). Pain scores as the verbal rating scale (VRS) or numerical rating scale (NRS) are used to determine pain management. While it is crucial to measure pain levels, it is equally important to identify patients who desire pain medication, so that adequate provision of analgesia can occur.Objective: To establish the association between pain scores on the NRS and VRS, and the desire for, and provision of, pain medication.Design, settings and participants: Retrospective monocentric observational cohort study of ED patients presenting with painful conditions.Outcomes measure and analysis: The primary outcome was to establish for each pain score (NRS and/or VRS), those patients who desired, and were ultimately provided with, pain medication, and those who did not. Secondary outcomes included establishing the prediction of pain scores to determine desire of pain medication, and the correlation between NRS and VRS when both were reported.Main Results: 130,279 patients were included for analysis. For each patient who desired pain medication, pain medication was provided. Proportion of patients desiring pain medication were 4.1–17.8% in the pain score range 0.5–3.5, 31.9–63.4% in the range 4–6.5, and 65–84.6% in the range 7–10. The prediction probability of pain scores to determine desire for pain medication was represented with an AUROC of 0.829 (95% CI 0.826–0.831). The optimal threshold predicting the desire for pain medication would be a pain score of 4.25, with sensitivity 0.86, and specificity 0.68. For the 7835 patients with both NRS and VRS scores available, the Spearman-Rho coefficient assessing correlation was 0.946 (p < 0.001).Conclusions: Despite guidelines currently recommending pain medication in patients with a NRS score > 4, we found a discrepancy between pain scores and desire for pain medication. Results of this large retrospective cohort support that the desire for pain medication in the ED might not be derived from a pain score alone.