This study compared recalled average pain, assessed at the end of the day, with the average of real-time pain ratings recorded throughout the day among patients with osteoarthritis (OA). Participants (N = 157) with hand, hip, or knee OA completed electronic pain diaries on 1 weekend day and 1 weekday. Diaries included at least 7 pain ratings per day, immediately after waking and every 2 hours following, using a visual analog scale (VAS) scored as 1 to 100 (scores not seen by participants). At the end of each diary day, participants rated their average pain that day on the same VAS. Pearson correlations examined associations between recalled average pain and the average of real-time pain ratings that day. Mixed models, including interaction terms, examined whether associations between recalled and actual average pain ratings differed according to the following patient characteristics: joint site, age, race, gender, study enrollment site, and pain catastrophizing. Correlations between recalled and actual average pain ratings were r = .88 for weekdays and r = .86 for weekends (P < .0001). In mixed models, there were no significant interaction terms for any patient characteristics. In summary, patients with OA accurately recalled their average pain over a 1-day period, and this did not differ according to any patient characteristics examined.
Perspective: This study showed that patients with OA accurately recalled their average pain over a single-day period, and this did not differ according to patient characteristics. Results of this study indicate that end-of-day recall is a practical and valid method for assessing patients' average pain during a day.
Published by Elsevier Inc.