The visual analog scale in the immediate postoperative period: Intrasubject variability and correlation with a numeric scale

The visual analog scale (VAS) has been used to assess the efficacy of pain management regimens in patients with acute postoperative pain, but its usefulness has not been confirmed in postoperative pain studies. We studied 60 subjects in the immediate postoperative period. The specific data collected were: VAS scores versus an Ii-point numeric pain scale; repeatability in VAS scores over a short time interval; and change in VAS scores from one assessment period to the next versus a verbal report of change in pain. The correlation coefficients for VAS scores with the Ii-point pain scale were 0.94, 0.91, and 0.95. The repeatability coefficients were 17.6, 23.0, and 13.5 mm. Of the 56 patients who completed all three assessments, only 16 (29%) had repeatability within 5 mm on all three. Some of the changes in VAS scores between assessments were in the direction opposite the verbally reported changes in pain (31%); however, most (92%) were within 20 mm. There was no correlation between the level of sedation, previous pain experience, anxiety, or anticipated pain with consistency in VAS scores. We conclude that any single VAS score in the immediate postoperative period should be considered to have an imprecision of +/-20 mm. Implications: The visual analog scale was developed for assessing chronic pain but is often used in studies of postoperative pain. This study finds that the visual analog scale correlates well with a verbal Ii-point scale but that any individual determination has an imprecision of +/-20 mm.