Background: The Animated Activity Questionnaire (AAQ) is an online, self-reported, animated questionnaire measuring how patients with hip or knee osteoarthritis (OA) perform activities of daily living (ADL. By means of animations of an avatar the AAQ displays several ADL tasks in 3-5 different levels of difficulty. Patients select the animation that best matches their performance of the activity in the past week ( https://animated activityquestionnaire.com ). So far, it is unclear whether the AAQ is suitable for monitoring recovery in patients with OA who underwent total hip or knee arthroplasty (THA or TKA).

Objectives: This study aimed to assess the construct validity and responsiveness of the AAQ in THA/TKA patients.

Methods: Participants in a nationwide, cluster randomized controlled trial comparing an optimized physical therapy intervention with usual care completed the AAQ at baseline and 3 and 12 months postoperatively (Score range 0-100), with 100 indicating the best possible score). To assess construct validity, correlations between the baseline AAQ and comparators were computed. Comparators were three performance based tests (30 seconds Chair-Stand Test, 40m (4x10m) Fast Paced Walk Test and Stair Climb Test) and two questionnaires (7-item short versions of the Hip Disability/Knee Injury and Osteoarthritis Outcome Scores and Oxford Hip or Knee Scores. Construct validity was assessed by means of testing 5 hypotheses about expected correlations between the AAQ and the comparators (expected correlations for all 5 moderate, Spearman’s r 0.3-0.6). Responsiveness was assessed by means of testing 20 hypotheses (10 for 3 months and 10 for 12 months) about expected correlations between change in AAQ and change in the scores of the comparators (expected correlations Spearman’s r >0.5 for 2 hypotheses, r >0.3 for 12 hypotheses, r more than 0.1 higher for correlations AAQ with specific comparators as compared with others for 6 hypotheses). We considered the AAQ as valid and/or responsive for THA and TKA separately if at least 75% of the results were in accordance with the hypotheses.

Results: 582 patients (302 THA, 282 TKA) were included in the analysis of the construct analysis, whereas 552 were included in the responsiveness analysis at 3 months and 460 at 12 months postoperatively. In the total THA and TKA groups, 52% and 55% were female, the mean age was 66.6 (SD 9.0) and 66.9 (SD 8.4) years and the mean AAQ score at baseline was 77.8 (SD 16.8) and 78.2 (SD 16.0), respectively. For all 5 hypotheses on the construct validity in both THA and TKA patients moderate correlations were found in both THA and TKA patients (correlations ranged between 0.51-0.62; 100% hypotheses met). With regard to the responsiveness analysis, the 2 correlations at 3 and 12 months where r>0.5 was expected were neither met in THA nor in TKA (r 0.19-0.33). All 12 hypotheses where r >0.3 was expected at 3 or 12 months were met in both THA and TKA (r 0.37-0.70), whereas 2 (in THA)and 3 (in TKA) of the 6 hypotheses where 0.1 higher correlations with specific comparators were expected at 3 or 12 months were met. Thereby, 70% (14 out of 20) of the hypotheses were met in THA and 75% (15 out of 20) in TKA patients at 3 and 12 months follow up.

Conclusions: Based on our results, the AAQ can assess physical function accurately in THA and TKA patients. For TKA patients we found sufficient responsiveness until 1 year follow-up, however, for THA patients responsiveness of the AAQ was limited. Nevertheless, the AAQ can be a useful tool in orthopaedic practice to assess ADL.

REFERENCES: NIL.

Acknowledgments: NIL.

Disclosure of Interests: None declared.