Differences in Clinically Significant Changes in Children's Pain Intensity Scores Based on Patient Characteristics

Objectives: To determine the minimum and ideal clinically significant differences (MCSD, ICSD) of the Faces Pain Scale--Revised (FPS-R) and the Color Analog Scale (CAS) in children, and to identify any differences in these estimates based on patient characteristics. Methods: We conducted a prospective study of children aged 4 to 17 years with acute pain presenting to two urban pediatric emergency departments. Participants self-reported their pain severity using the FPS-R and CAS, and qualitatively described their change in pain. Changes in pain score reported using the FPS-R and CAS that were associated with ``a little less'' and ``much less'' pain (MCSD and ICSD, respectively) were identified using a receiver operating characteristic-based method, and expressed as raw change score and percent reductions. Estimates of MCSD and ICSD were determined for each category of initial pain severity (i.e. mild, moderate, and severe), and patient characteristics (age, sex, and ethnicity). Post-hoc exploratory analyses evaluated categories of race, primary language, and etiology of pain. Results: 314 children were enrolled; mean age was 9.8 years. The FPS-R raw change score and percent reduction MCSD estimates were 2/10 and 25%, with ICSD estimates of 3/10 and 60%. For the CAS, raw change score and percent reduction MCSD estimates were 1/10 and 15%, with ICSD estimates of 2.75/10 and 52%. For both scales, raw change score and percent reduction estimates of the MCSD remained unchanged in children with either moderate or severe pain. For both scales, estimates of ICSD were not stable across categories of initial pain severity. There was no difference in MCSD or ICSD based on age, sex, ethnicity, race, primary language, or etiology of pain. Conclusions: We have established MCSD estimates expressed as raw change score and percent reductions for the FPS-R and CAS. These estimates appear stable for children with moderate to severe pain, irrespective of age, sex, and ethnicity. Estimates of ICSD were not stable across different categories of initial pain severity, therefore limiting their potential generalizability.