Computer-aided surface estimation of pain drawings – intra- and inter-rater reliability

• Main Authors: Persson AL, Garametsos S, Pedersen J
• Format: Article
• Language: English
• Published: Dove Medical Press 2011-05-01
• Series: Journal of Pain Research
• Online Access: http://www.dovepress.com/computer-aided-surface-estimation-of-pain-drawings-ndash-intra--and-in-a7468

Ann L Persson¹, Sofia Garametsos²,*, Jonna Pedersen²,* ¹Rehabilitation and Research Centre for Torture Victims, Copenhagen K, Denmark; ²Department of Health Sciences, Division of Physiotherapy, Lund University, Lund, Sweden *These authors contributed equally to this studyAbstract: Pain drawings are often utilized in the documentation of pain conditions. The aim here was to investigate intra- and inter-rater reliability of area measurements performed on pain drawings consecutively, using the computer program Quantify One. Forty-eight patients with chronic nonmalignant pain had shaded in their experienced pain on the front and back views of a pain drawing. The templates were scanned and displayed on a 17-inch computer screen. Two independent examiners systematically encircled the shaded-in areas of the pain drawings with help of a computer mouse, twice each on two separate days, respectively. With this method it is possible to encircle each marked area and to obtain immediate details of its size. The total surface area (mm2) was calculated for each pain drawing measurement. Each examiner measured about 2400 areas, and as a whole, the number of areas measured varied only by 3%. The intra-rater reliability was high with intraclass correlation coefficients 0.992 in Examiner A and 0.998 in Examiner B. The intra-individual absolute differences were small within patients within one examiner as well as between the two examiners. The inter-rater reliability was also high. Still, significant differences in the absolute mean areas (13%) were seen between the two examiners in the second to fourth measurement sessions, indicating that one of the examiners measured systematically less. The measurement error was ≤10%, indicating that use of the program would be advantageous both in clinical practice and in research, but if repeated, preferably with the same examiner. Since pain drawings with this method are digitized, high quality data without loss of information is possible to store in electronic medical records for later analysis, both regarding precise location and size of pain area. We conclude that the computer program Quantify One is a reliable method to calculate the areas of pain drawings.Keywords: area calculation, assessment, chronic pain, digitized pain drawing