Pain Assessment in the Emergency Department: A Prospective Videotaped Study

Deep Learning-Based Instance Appraisable Model (EDi Pain) for Pain Estimation via Facial Videos: A Retrospective Analysis and a Prospective Emergency Department Study

Pain assessment is a critical aspect of medical care, yet automated systems for clinical pain estimation remain rare. Tools such as the visual analog scale (VAS) are commonly used in emergency departments (EDs) but rely on subjective self-reporting, with pain intensity often fluctuating during triage. An effective automated system should utilize objective labels from healthcare professionals and identify key frames from video sequences for accurate inference. In this study, short video clips were treated as instance segments for the model, with ground truth (physician-rated VAS) provided at the video level. To address the weak label problem, we proposed flexible multiple instance learning approaches. Using a specialized loss function and sampling strategy, our instance-appraisable model, EDi Pain, was trained to estimate pain intensity while evaluating the significance of each instance segment. During inference, the VAS pain score for the entire video is derived from instance-level predictions. In retrospective analysis using the public UNBC-McMaster dataset, the EDi Pain model demonstrated competitive performance relative to prior studies, achieving strong performance in video-level pain intensity estimation, with a mean absolute error (MAE) of 1.85 and a Pearson correlation coefficient (PCC) of 0.63. Additionally, our model was validated on a prospectively collected dataset of 931 patients from National Taiwan University Hospital, yielding an MAE of 1.48 and a PCC of 0.22. In summary, we developed and validated a novel deep learning-based, instance-appraisable model for pain intensity estimation using facial videos. The EDi Pain model shows promise for real-time application in clinical settings, offering a more objective and dynamic approach to pain assessment.

Prospective study of pain and patient outcomes in the emergency department: a tale of two pain assessment methods

BACKGROUND

Accurate pain assessment is essential in the emergency department (ED) triage process. Overestimation of pain intensity, however, can lead to unnecessary overtriage. The study aimed to investigate the influence of pain on patient outcomes and how pain intensity modulates the triage's predictive capabilities on these outcomes.

METHODS

A prospective observational cohort study was conducted at a tertiary care hospital, enrolling adult patients in the triage station. The entire triage process was captured on video. Two pain assessment methods were employed: (1) Self-reported pain score in the Taiwan Triage and Acuity Scale, referred to as the system-based method; (2) Five physicians independently assigned triage levels and assessed pain scores from video footage, termed the physician-based method. The primary outcome was hospitalization, and secondary outcomes included ED length of stay (EDLOS) and ED charges.

RESULTS

Of the 656 patients evaluated, the median self-reported pain score was 4 (interquartile range, 0-7), while the median physician-rated pain score was 1.5 (interquartile range, 0-3). Increased self-reported pain severity was not associated with prolonged EDLOS and increased ED charges, but a positive association was identified with physician-rated pain scores. Using the system-based method, the predictive efficacy of triage scales was lower in the pain groups than in the pain-free group (area under the receiver operating curve, [AUROC]: 0.615 vs. 0.637). However, with the physician-based method, triage scales were more effective in predicting hospitalization among patients with pain than those without (AUROC: 0.650 vs. 0.636).

CONCLUSIONS

Self-reported pain seemed to diminish the predictive accuracy of triage for hospitalization. In contrast, physician-rated pain scores were positively associated with longer EDLOS, increased ED charges, and enhanced triage predictive capability for hospitalization. Pain, therefore, appears to modulate the relationship between triage and patient outcomes, highlighting the need for careful pain evaluation in the ED.