Abstract:
OBJECTIVES: This study examined whether risk factors for knee osteoarthritis (KOA) pain such as age, gender, body mass index (BMI), baseline pain, and other putative risk factors for knee osteoarthritis pain flares (KOAF) (e.g. knee buckling, injury, mood/stress/social support scores, and footwear) could predict KOAF. METHOD: People with KOA and previous history of KOAF were selected from a 3-month web-based longitudinal study. KOAF was defined as an increase of ≥ 2 points on a numeric rating scale (compared with background pain) which resolved within 20 days. Predictors assessed at baseline were gender, age, duration of KOA, BMI, pain, knee injury (7 days before), knee buckling (2 days before), Lubben Social Support, Knee Injury and Osteoarthritis Outcome Score, Intermittent and Constant Osteoarthritis Pain score (ICOAP), Positive/Negative Affect Score, and footwear stability/heel height. Outcome was occurrence of any KOAF during the ensuing 30 days. The combined ability of the above variables to predict occurrence of any KOAF was evaluated by multiple logistic regression with a 10-fold cross-validation method to build and internally validate the model. Variables that assessed similar domains were eliminated using receiver operating characteristics curve assessment for best fit. RESULTS: Complete data were available for 313 people (66.6% female, mean ± sd age 62.3 ± 8.2 years, BMI 29.7 ± 6.5 kg/m2). Increasing age, years of osteoarthritis, BMI, background/worst levels of pain, knee injury, knee buckling, ICOAP, and footwear category/heel height significantly predicted the occurrence of KOAF during the following 30 days, with an area under the curve of 0.73 (95% confidence interval 0.67-0.80). Conclusion: A combination of risk factors assessed at baseline, including exposures with potential to vary, successfully predicts the KOAF in the ensuing 30 days.