Pain outcome measures and clinically important change

Measure pain in several domains, not intensity alone, and judge improvement by percentage: at least 30% is moderately important and at least 50% substantial for an individual, while between-group differences need separate judgment.

1. IMMPACT benchmarks four core outcome domains. Pain intensity on a 0–10 numerical rating scale; physical functioning by the Multidimensional Pain Inventory or Brief Pain Inventory interference scales; emotional functioning by the Beck Depression Inventory or Profile of Mood States; and global improvement ¹.
2. Global change is patient-rated. IMMPACT uses the Patient Global Impression of Change to capture participants' ratings of overall improvement, the anchor against which pain-score changes are calibrated ¹².
3. Percentage change travels best. About a 2-point or 30% fall on the 0–10 scale matched patients rating themselves "much improved," consistently across conditions and regardless of baseline pain ².
4. Use the IMMPACT benchmarks. A reduction of at least 30% is moderately important and at least 50% substantial; IMMPACT also advises using two or more methods to judge clinical importance ¹.
5. The PEG is a brief primary-care tool. Three items from the BPI cover average pain intensity, interference with enjoyment of life and interference with general activity; the full BPI is widely used in specialty and research settings but is impractical for primary care ³.
6. Between-group thresholds differ. Applying within-patient benchmarks to active-versus-placebo differences is a common misinterpretation; IMMPACT says such differences need case-by-case evaluation ⁴.

Higher baseline pain requires larger raw point changes to be clinically important, so a fixed point change misleads when baseline scores differ; percentage change related consistently to global improvement across baselines ².

• Farrar pooled 2724 patients from 10 pregabalin trials across five conditions ².
• PEG reliability was alpha 0.73 and 0.89 in two samples, with construct validity comparable to the BPI ³.
• The PEG was sensitive to change and separated improvers from non-improvers at 6 months ³.
• Single-item pain screening gives limited information about chronic pain ³.
• Trivial between-group differences can become statistically significant in large trials, so magnitude needs separate judgment ⁴.
• STarT Back used the Roland Morris Disability Questionnaire at 12 months as its primary outcome; stratified care improved it by 1.06 points versus best usual care ⁵.
• PHQ-9 scores of 10 or more identify major depression with 88% sensitivity and specificity ⁶.

• Judging response by a fixed 2-point drop regardless of baseline ².
• Reporting pain intensity alone as the trial outcome ¹.
• Dismissing a between-group difference because it is below 30% ⁴.
• Relying on a single-item pain score to monitor chronic pain in primary care ³.