Knee confidence trajectories over 8 years and factors associated with poor trajectories: data from the osteoarthritis initiative

Abstract

Purpose: Lack of knee confidence, one's perception of not being able to rely on his or her knees, is a frequent complaint and has been associated with function decline in persons with or at higher risk for knee osteoarthritis (OA). Given the central role of knee in weight-bearing activities, confidence in knees could influence not only physical function but also activity participation. A better understanding of how confidence changes over time and identification of modifiable factors associated with poor trajectories will inform strategies to prevent function decline and promote activity participation. Our objectives were to identify, in persons with or at higher risk for knee OA, the number of distinct trajectories of knee confidence over 8 years and baseline factors associated with poor trajectories. Methods: The Osteoarthritis Initiative (OAI) is a prospective longitudinal cohort study of persons with or at higher risk for knee OA, aged 45–79 years, recruited from 4 communities. Knee confidence was self-reported annually from baseline to 8 years (up to 9 time points) in the KOOS (Knee Injury and Osteoarthritis Outcome Score) Knee-related Quality of Life subscale. Participants responded to the question: How much are you troubled with lack of confidence in your knees, on a 5-point Likert scale of not at all (0), mildly (1), moderately (2), severely (3), and extremely (4). Lack of knee confidence was defined as scores of ≥ 2. Using available data over 8-year follow-up period, in 4515 OAI participants who had baseline and ≥ 2 follow-up time points for confidence score, latent class models identified subgroups that shared a similar underlying knee confidence trajectory for probability of reporting lack of knee confidence. To select the number of trajectory subgroups, we used Bayesian Information Criteria (BIC), modeled using intercept only, linear, quadratic, and cubic terms. We varied the number of trajectory subgroups from 2 to 5 until the best-fit model emerged. In 4105 persons with complete baseline data, we used logistic regression to assess the association between each of the baseline predictors and poor (vs. good) trajectories. Results: The analysis sample had a mean age of 61.2 years (SD 9.2), BMI 28.6 kg/m2 (4.8), and 2640 (58.5%) were women. We identified 4 distinct knee confidence trajectories (Figure), persistently good, declining, improving, and persistently poor. Among 4515 participants, 66% were in the persistently good trajectory, 9% in the declining, 14% in the improving, and 11% in the persistently poor. As shown in the Table, baseline factors associated with both persistently poor and declining confidence trajectories were BMI, depressive symptoms, disease severity, and worse function; with persistently poor only were younger age, male sex, knee extensor weakness, history of knee injury, and knee and ankle pain; with declining only were comorbidities, falls, and hip pain. Conclusions: Four distinct 8-year knee confidence trajectories were identified in persons with or at higher risk for knee OA. Reducing BMI, depressive symptoms, and knee, ankle, and hip pain as well as improving function and extensor strength may potentially alter the knee confidence trajectory. Findings suggest the need for multi-modal intervention studies to maintain or improve knee confidence.View Large Image Figure ViewerDownload Hi-res image Download (PPT)