Predicting the course of functional limitation among older adults with knee pain: do local signs, symptoms and radiographs add anything to general indicators?

OBJECTIVE: To determine the additional prognostic value of clinical history, physical examination and x-ray findings to a previously derived simple generic model (age, body mass index, anxiety and pain severity) in a cohort of older adults with knee pain.
METHODS: Prospective cohort study in community-dwelling adults in North Staffordshire. 621 participants (aged >or=50 years) reporting knee pain who attended a research clinic at recruitment and were followed up by postal questionnaire at 18 months. Poor functional outcome was measured by the Physical Functioning Scale of the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) at 18-month follow-up defined in 60% of participants.
RESULTS: Three clinical history variables (bilateral knee pain, duration of morning stiffness and inactivity gelling) were independently associated with poor outcome. The addition of the "clinical history" model to the "generic" model led to a statistical improvement in model fit (likelihood ratio (LR) = 24.84, p = 0.001). Two physical examination variables (knee tender point count and single-leg balance) were independently associated with poor outcome but did not lead to a significant improvement when added to the "clinical history and generic" model (LR = 6.34, p = 0.50). Functional outcome was significantly associated with severity of knee radiographic osteoarthritis (OA), but did not lead to any improvement in fit when added to the "generic, clinical history and physical examination" model (LR = 1.86, p = 0.39).
CONCLUSIONS: Clinical history, physical examination and severity of radiographic knee OA are of limited value over generic factors when trying to predict which older adults with knee pain will experience progressive or persistent functional difficulties.