Osteoarthritis (OA) is a leading cause of global disability. International guidelines make clear recommendations for evidence-based OA management. However, there is considerable discrepancy between these recommendations and the actual care received by patients. Osteoarthritis management programs (OAMPs) aim to address this evidence-practice gap. There is evidence that some participants improve in pain and function following OAMPs, however, others fail to accomplish these gains. The ability to predict patient outcomes would enable targeting these programs at those people most likely to demonstrate improvement. This Thesis addresses the question: ‘Can clinical presentation can predict response to an OAMP?’ Five studies were conducted to address this question.
Two longitudinal cohort studies examined the relationships between participant characteristics and changes in pain and function following 26 weeks of an OAMP. Significant predictors included: sex; knee as treatment joint (vs hip); and total joint arthroplasty (TJR) waitlist status. However, the regression models were not sufficiently sensitive to correctly classify ‘responders’ or ‘worseners’.
We then examined patients’ attitudes and capabilities towards OA self-management as a construct that could potentially predict OAMP outcomes. We conducted a systematic review that aimed to identify the instrument assessing OA self-management attitudes and capabilities with the “best” measurement properties. From this review, little extant measurement property evidence was found to recommend any instrument.
The fourth study examined the measurement properties of the Patient Activation Measure (PAM-13) in a cross-sectional cohort study. The fifth study examined the relationships between PAM-13 and changes in pain and function. PAM-13 scores were not associated with changes in pain or function following 12 or 26 weeks of the OAMP. Other variables that were independent predictors of outcomes were the Timed-Up-and-Go and employment status.
It is difficult to determine who will improve or worsen in an OAMP. This Thesis did identify some patient characteristics that were predictors of response, however, these findings should be replicated in larger cohorts.