Background/Purpose: Up to 20% of Total Joint Replacement (TJR) patients are dissatisfied, but this is difficult to study as it is challenging to pool data due to the lack of unified core outcome measures.  The OMERACT TJR Working Group has recently endorsed a core domain set for Knee and Hip TJR trials that include pain, function and patient satisfaction among others, and now seeks validation prior to development of a TJR trial core measurement set. We aim to assess the association of pain relief and improved function with patient satisfaction 2 years after TJR.

Methods:  We identified all patients undergoing total hip (THR) and knee (TKR) replacement enrolled in a hospital-based registry from 2007-2011, and evaluated those with 2-year satisfaction scores. Pain and function were measured using the Knee and Hip injury and Osteoarthritis Outcome Score (KOOS, HOOS) and satisfaction was measured using 5 primary questions, each rated on a Likert scale.  Each question was weighted equally and a satisfaction summary score was calculated (range 0-100, higher scores corresponding to greater satisfaction). Expectations were measured using the validated HSS Expectations survey. Correlation was analyzed with Spearman coefficients, and scores were compared by quartiles using the Kruskal-Wallis test. Rasch modeling was attempted but a unidimensional construct could not be achieved.

Results: We included 4,796 primary unilateral TKR and 4801 THR. 78% of TKR and 90.7% of THR were very satisfied with pain relief, and 6.5% of TKR and 2.5 % of THR were somewhat or very dissatisfied (Table 1). Satisfaction correlated moderately with pain (TKR ρ=0.61, THR ρ=0.47) and function (TKR ρ=0.65, THR ρ=0.51) at 2 years; there was no correlation with baseline expectations. When comparing satisfaction by pain, function and expectation quartiles, there were statistically significant differences (Table 2); those with the best scores and greatest change in pain and function were the most satisfied.

Conclusion: These findings confirm that with increasing relief of pain and functional improvement, the strength of the association of 2 core domains with satisfaction increases, further validating these core domains for use in TJR clinical trials. The range of correlation of satisfaction measure of 0.47-0.61 with pain and function, indicates that satisfaction domain is somewhat independent of pain and function, further validating its inclusion in the core domain set. A core outcome measurement set needs to be defined for use in TJR clinical trials that includes validated measures of these domains.