A prospective population-based study of the predictors of undergoing total joint arthroplasty

Objective. To examine prospectively the predictors of time to total joint arthroplasty (TJA). Methods. This was a prospective cohort study with a median followup time of 6.1 years. We included participants from an existing population-based cohort of 2,128 individuals, ages 55 years and older with disabling hip and/or knee arthritis and no prior TJA, from 2 regions of Ontario, Canada, 1 urban with low TJA rates and 1 rural with high rates. The main outcome measure was the occurrence of a TJA based on procedure codes in the hospital discharge abstract database. Results. At baseline, the mean age of the patients was 71.5 years, 67.9% had a high school education or higher, 73.4% were women, the mean arthritis severity (Western Ontario and McMaster Universities Osteoarthritis Index [WOMAC] score was 41.1 (maximum possible score 100), and 20.0% were willing to consider TJA. Greater probability of undergoing TJA was associated with higher (worse) baseline WOMAC scores (hazard ratio [HR] 1.22 per 10-unit increase, P < 0.001), age (compared with age <= 62 years, the HR increased to 1.57 for 63-68 years, 1.46 for 69-74 years, and 1.51 for 75-81 years, and fell to 0.44 for >= 82 years; P < 0.05 for all), better health (HR 1.14 per 10-unit increase in Short Form 36 general health survey score, P < 0.001), and willingness to consider TJA (HR 4.92, P < 0.001). When willingness was excluded from the model, education level, but not sex or income, became a significant predictor of TJA receipt. Conclusion. Willingness to consider TJA was the strongest predictor of the time to first TJA. Given that previous research indicates that willingness is largely explained by perceptions of the indications for and risks associated with TJA and not disease severity, this finding supports the need for population education about arthritis treatments, including TJA.