To compare the accuracy of daily and recalled pain measurements in hip and knee osteoarthritis (OA).

A prospective study investigating pain intensity over 29 days in patients with painful OA. Pain was assessed on days 1 and 29 during visits. Between these two visits, daily (worst, least, usual, mean, at rest and during daily life) and recalled pain intensities were randomly recorded by telephone interview with a numerical rating scale.

We studied 129 patients, with a mean age of 67.7±10.0 years, 27.1% males. Daily prospective assessments demonstrated that the strongest correlations were between ‘mean daily’ and ‘usual’ pain (r=0.88) and between ‘mean daily’ and ‘worst’ pain (r=0.86). Retrospective assessments demonstrated a strong correlation between recalled pain intensities and calculated mean of daily assessments over the seven and 28 days (0.78 and 0.67, respectively), but weakened by adjustment for pain intensity on the day of recalled pain assessment. Anxiety and depression scores did not affect pain recall for 7 and 28 days.

In hip and knee OA, prospective daily pain assessment can be performed either by assessing ‘usual pain’ or ‘worst pain’ of the day. When recalled over one and four weeks, recall pain intensity is well correlated with calculated mean of prospective daily pain assessments during the same period, but correlations are weaker as the length of period increases. Either for 1 and 4 weeks, retrospective assessments are influenced by current pain intensity at the end of the period.