Reliability and concurrent validity of the PainMAP for Automated Quantification of Pain Drawings in Body Charts in Patients with Low Back Pain

Introduction: Low back pain (LBP) is the primary cause of years lived with disability in the world1,2. Widespread pain was observed in 24% of patients with LBP3 and patients with widespread pain that includes low back area have more comorbidities, psychosocial factors, pain duration, and functional limitations than patients with local LBP4. The assessment of painful areas through printed body charts is a simple way to clinicians identifies patients with widespread pain in primary care5. However, there is a lack in the literature about a simple and automated method designed to analyze pain drawings in body charts in clinical practice. Objective: To test the inter- and intra-rater reliability and concurrent validity of a software (PainMAP) for quantification of pain drawings in patients with low back pain. Methods: Thirty-eight participants [16(42.10%) female; mean age 50.24(11.54) years; mean body mass index 27.90(5.42) kg/m2; duration of pain of 94.35(96.11) months] with a current episode of low back pain were recruited at physiotherapy outpatients. This study was approved by the Research Ethics Committee of Federal Institute of Rio de Janeiro and was conducted in accordance with the Declaration of Helsinki of 1975. All subjects included in the study signed the Informed Consent Form. Participants were instructed to shade all their painful area in a body chart using a red pen. The body charts were digitized by separate raters using smartphone cameras and twice for one rater to analyze the intra-rater reliability. Both the number of pain sites and the pain area were calculated using ImageJ (reference method). The PainMAP software used image processing methods to automatically quantify the data from the same digitized body charts. Results: The reliability analyses revealed that the PainMAP has excellent inter- and intra-rater reliability to quantify the number of pain sites [ICC2,1: 0.998 (95%CI 0.996-0.999); ICC2,1: 0.995 (95%CI 0.991-0.998)] and pain area [ICC2,1: 0.998 (95%CI 0.995-0.999); ICC2,1: 0.975 (95%CI 0.951-0.987)], respectively. The standard error of the measurement was 0.22 (4%) for the number of pain sites and 0.03 (4%) cm2 for the pain area. The Bland-Altman analyses revealed no substantive differences between the two methods for pain area [mean difference = 0.007 (95%CI 0.067-0.053)]. Conclusion: The PainMAP software is reliable and valid to be used for quantification of the number of pain sites and pain area in patients with low back pain. Keywords: Low back pain; Reliability; Pain Measurement.