• Marco Barbero, Corrado Cescon, Alessandro Schneebeli, Deborah Falla, Giuseppe Landolfi, Marco Derboni, Vincenzo Giuffrida, Andrea Emilio Rizzoli, Paolo Maino, and Eva Koetsier.
• Rehabilitation Research Laboratory 2rLab, Department of Business Economics, Health and Social Care, University of Applied Sciences and Arts of Southern Switzerland, Manno, Switzerland (M.B., C.C., A.S.). Electronic address: marco.barbero@supsi.ch.
• J Pain Symptom Manage. 2024 Feb 1; 67 (2): e129e136e129-e136.

IntroductionPen-on-paper pain drawing are an easily administered self-reported measure that enables patients to report the spatial distribution of their pain. The digitalization of pain drawings has facilitated the extraction of quantitative metrics, such as pain extent and location. This study aimed to assess the reliability of pen-on-paper pain drawing analysis conducted by an automated pain-spot recognition algorithm using various scanning procedures.MethodsOne hundred pain drawings, completed by patients experiencing somatic pain, were repeatedly scanned using diverse technologies and devices. Seven datasets were created, enabling reliability assessments including inter-device, inter-scanner, inter-mobile, inter-software, intra- and inter-operator. Subsequently, the automated pain-spot recognition algorithm estimated pain extent and location values for each digitized pain drawing. The relative reliability of pain extent analysis was determined using the intraclass correlation coefficient, while absolute reliability was evaluated through the standard error of measurement and minimum detectable change. The reliability of pain location analysis was computed using the Jaccard similarity index.ResultsThe reliability analysis of pain extent consistently yielded intraclass correlation coefficient values above 0.90 for all scanning procedures, with standard error of measurement ranging from 0.03% to 0.13% and minimum detectable change from 0.08% to 0.38%. The mean Jaccard index scores across all dataset comparisons exceeded 0.90.ConclusionsThe analysis of pen-on-paper pain drawings demonstrated excellent reliability, suggesting that the automated pain-spot recognition algorithm is unaffected by scanning procedures. These findings support the algorithm's applicability in both research and clinical practice.Copyright © 2023 The Authors. Published by Elsevier Inc. All rights reserved.

15 keywords...

hide…