Description
Objectives
The aim this study is to quantify the presence of recall bias in the evaluation of knee pain and function in knee osteoarthritis (OA) and to identify its determinants to prevent a biased symptom interpretation.
Methods
This prospective multicentric study enrolled 115 patients with knee OA who used a mobile App for their smartphone to collect punctual pain and function on two 0-10 numerical rating scales (NRS) 2 times a day for 2 months. At the 1-month and at the 2-month follow-up visits, patients were asked to retrospectively evaluate the mean level of pain and function of the last month on a 0-10 NRS. Recall bias was computed as the difference between the mean level of pain/function reported using the App and the level of pain/function reported with the retrospective assessment.
Results
Recall bias documenting an overestimation of symptoms with the retrospective assessment was documented for pain at the first (MD=-0.24; p=0.04) and at the second (MD=-0.64; p<0.001) month of follow-up. Recall bias affected also function assessment both at the first (MD=0.96; p<0.001) and at the second (MD=0.80; p=0.001) month of follow-up. A longer duration of symptoms, greater baseline symptoms in terms of both pain and function NRS, and symptoms worsening during the month before the evaluation were all significantly correlated with a higher recall bias.
Conclusions
The retrospective assessment of pain and function in knee OA, as commonly performed in the clinical practice and in research studies, is significantly affected by recall bias. Patients reporting worse outcomes are more prone to recall bias and part of their symptoms may be overestimated. Daily assessment of symptoms with a smartphone may improve patient management and research outcome assessment in knee OA.
