Pain Without Words: Emoji-Based Assessment of Acute Pain in the Emergency Department—Clinical and Medico-Legal Impact

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background: Pain is among the most frequent reasons for seeking medical care and presenting to the Emergency Department (ED). In Italy, Law No. 38/2010 establishes the right to pain therapy and requires pain intensity and its evolution during treatment to be assessed and documented. This process may be challenging in patients with language barriers because conventional verbal or numerical instruments require a degree of linguistic or numerical comprehension. Emoji-based visual tools may provide an additional means of communicating pain intensity in linguistically diverse clinical settings. Methods: We conducted a single-center, prospective, observational pilot study involving adult patients presenting to the ED with acute pain and a clinically relevant language barrier. Pain was assessed before and after analgesic treatment using a six-category Emoji-Based Pain Scale, scored 0, 2, 4, 6, 8, and 10, and the 11-point Numeric Rating Scale (NRS). Feasibility, independent completion, monotonic association, and agreement between the instruments were evaluated. Spearman’s rank correlation, exploratory Bland–Altman analysis, and quadratic-weighted Cohen’s kappa were calculated separately before and after treatment. A post hoc ancillary rank-ordering exercise was subsequently conducted in 50 adult volunteers using the same unlabelled and randomly shuffled emoji cards. Results: Eighty-seven ED patients were enrolled. The Emoji-Based Pain Scale was completed independently by 78 patients (89.7%), compared with 61 patients (70.1%) for the NRS. Mean pre-treatment scores were 4.90 ± 1.77 for the Emoji-Based Pain Scale and 5.15 ± 1.65 for the NRS; post-treatment scores were 2.16 ± 1.62 and 2.46 ± 1.73, respectively. Strong positive associations were observed before treatment (Spearman’s ρ = 0.860, p < 0.001) and after treatment (ρ = 0.805, p < 0.001). Mean paired differences were −0.25 points before treatment and −0.30 points after treatment. Quadratic-weighted kappa was 0.747 before treatment (95% bootstrap confidence interval [CI], 0.661–0.818) and 0.720 after treatment (95% bootstrap CI, 0.609–0.806). Exact categorical agreement was observed in 62.1% and 58.6% of assessments, respectively. In the ancillary exercise, all 50 volunteers reproduced the prespecified emoji sequence, with no inversion of the two upper anchors. Conclusions: The Emoji-Based Pain Scale was feasible and more frequently completed without assistance than the NRS in this selected population of adult ED patients with language barriers. Strong monotonic association, good ordinal agreement, and limited mean differences were observed between the instruments. These findings support further evaluation of the scale but do not establish formal validity, reliability, or interchangeability. Direct psychometric comparison with established visual pain scales remains necessary before routine clinical integration.