The value of remembered pre-operative quick disabilities of the arm, shoulder and hand (QuickDASH) scores.

It would be useful if it were possible for the patients to recreate their pre-operative QuickDASH scores in audits where this score had not been recorded before surgery. We assessed the accuracy of remembered pre-operative QuickDASH scores among 229 consecutive patients and the value of a previously developed algorithm for correcting these scores. Real pre-operative scores and remembered pre-operative scores were compared after a mean of 21 months. Furthermore, the scores of a subgroup of 79 patients with carpal tunnel syndrome, subacromial impingement, thumb basal joint arthrosis or Dupuytren's contracture were corrected using an algorithm. The mean difference between remembered and real pre-operative scores for all patients showed heteroscedacity in the Bland-Altman plot. The scores of the 79 sub-analysis patients were homoscedastic. The mean difference between remembered and real pre-operative scores was 9 (SD 16, SEM 1.85). Correcting the scores of the sub-group patients using our algorithm decreased the variation only moderately. The remembered pre-operative score is too inaccurate to be useful in individual patients, also when using our algorithm. However, subtracting nine from the mean remembered pre-operative score in a group of patients with any of the above diagnoses gives the real pre-operative score within the 95% confidence interval of four above and four below the real score.