Longitudinal Relationship Between Knee Pain Status and Incident Frailty: Data from the Osteoarthritis Initiative

Objective Examine the longitudinal association between knee pain and prefrailty/frailty. Design Longitudinal study. Setting Five clinical centers across the United States. Subject Data from 3,053 nonfrail participants aged 45–79 years at baseline from the Osteoarthritis Initiative. Methods According to self-reported knee pain at baseline, the participants were placed into three groups: no knee pain (N = 1,600), unilateral knee pain (N = 822), and bilateral knee pain (N = 631). Frailty status was assessed over time using the five frailty indicators (unintentional weight loss, exhaustion, weak energy, slow gait speed, and little physical activity). Based on the number of frailty indicators present, prefrailty (1–2) and frailty (≥3) were diagnosed. Generalized estimating equations logistic regression analyses were conducted to examine the relationship between knee pain status and prefrailty/frailty. Results After adjusting for age, sex, race, education, marital status, smoking status, comorbidities, and body mass index, unilateral knee pain at baseline was associated with an increased odds of developing prefrailty (odds ratio [OR] = 1.14, 95% confidence interval [CI] = 1.01–1.27) and frailty (OR = 1.89, 95% CI = 1.38–2.62), and bilateral knee pain at baseline was also associated with an increased risk of prefrailty (OR = 1.41, 95% CI = 1.24–1.62) and frailty (OR = 2.21, 95% CI = 1.63–3.01) over time in comparison with no knee pain. The interaction of knee pain status by time was not significantly associated with either prefrailty or frailty. Conclusions Knee pain (particularly bilateral knee pain) is associated with an increased risk of developing prefrailty and frailty over time.