Relationships between the knee injury and osteoarthritis outcome score (KOOS) and smartphone ecological momentary assessment outcomes in people with knee osteoarthritis.

Introduction: The Knee Injury and Osteoarthritis Outcome Score (KOOS) is a recommended recall-based self-report questionnaire assessing pain and function in people with knee osteoarthritis (OA). Smartphone Ecological Momentary Assessment (EMA) could be a more accurate method to assess current symptoms during everyday life. Whether outcomes collected via smartphone EMA align with recall-based patient-reported outcome measures (PROMs), such as the KOOS, is yet to be explored. Therefore, this study aims to explore relationships between outcomes collected via smartphone EMA and the KOOS in people with knee OA. Secondary aims include exploring whether relationships are maintained after adjusting for demographic and clinical factors influencing knee OA outcomes. Methods: A smartphone EMA survey was developed, piloted and then administered to people with knee OA. Participants were asked to rate current symptoms (i.e. pain intensity, pain interference, pain bothersomeness and stiffness) three times daily at random times during the morning, afternoon and evening for two weeks. Following this, participants completed a KOOS to collect pain, symptoms, function (daily living, sport and recreation) and quality of life outcomes. Bivariate correlations were performed to explore relationships between outcomes collected via smartphone EMA and the KOOS. Partial correlations were then performed to control for age, sex, body mass index (BMI) and baseline pain intensity. Results: Eighty-six people with knee OA completed both smartphone EMA and the KOOS. There was no loss to follow-up. Statistically significant inverse relationships (ranging in strength from weak to strong) were demonstrated between pain intensity (r = -0.69, p = <0.01), interference (r = -0.62, p = <0.01), bothersomeness (r = 0.78, p = <0.01) and knee stiffness (r = 0.61, p = <0.01) measured via the two methods. After controlling for established covariates, statistically significant inverse relationships were maintained although the strength of these relationships reduced overall, ranging from weak to moderate. Discussion and conclusion: Smartphone EMA appears to be valid when compared with recommended recall-based PROMs assessing pain and function in people with knee OA, even after controlling for knee OA prognostic factors. Collecting pain and function outcomes in a momentary manner could offer greater ecological validity to living with knee OA. Additionally, EMA could help to improve our understanding of concurrent psychosocial and lifestyle factors in shaping pain and functional outcomes in people with knee OA.