Purpose: Our prior work showed that achieving important improvement (II) in function by one year post total knee replacement (TKR) was directly associated with satisfaction with outcome. However, understanding the impact of the time of achievement of II has the potential to further our understanding of outcomes and inform care pathways. The objective of this work was to determine if the timing to achieve II in WOMAC pain and function and higher demand function as measured by the Late Life Disability (LLDI) was associated with satisfaction. Methods: We followed 354 patients from pre- to 1-year post-TKR. Pre-surgery, participants completed demographic and health questionnaires, the WOMAC pain and function and LLDI limitation subscales and identified outcome expectations for pain, other symptoms, mobility/activities of daily living (ADL), and participation in social roles/instrumental activities of daily living (IADL). The WOMAC and LLDI questionnaires also were completed at 3, 6, and 12 months post surgery. Satisfaction was completed at 12 months. All were scored 0-100 with higher scores indicating better outcome. Using II values for WOMAC pain, function and high demand activities generated from a separate cohort, we derived an ordinal variable of achieved II by 3, 6 or 12 months post-surgery or not achieved for each outcome. Based on sensitivity analysis, we categorized those who had baseline scores precluding achievement of II who reached the measure ceiling as achieving the II. We used a Bayesian path model with non-informative priors to evaluate if time to II achievement was associated with outcome satisfaction, adjusting for age, sex, education, obesity, depression (Hospital Anxiety and Depression Scale), comorbidity count and self-rated health. Expectations were modeled as individual predictors. As the Bayesian model (model 1) provided only a single estimator for our ordinal II variables, we also fit a generalized linear model to understand which time(s) of II achievement was associated with satisfaction (model 2). Results: Sample mean age was 65 years; 65% were female; and, 66% had >high school education. The mean comorbidity count was 1.6 (sd = 1.3) and depression was 5.3 (sd = 3.5). 40% rated their overall health as very good/excellent. The mean pre-surgery pain, function and high demand activities scores were 47.8 (sd = 17.8), 50.3 (sd = 18.5) and 59.2 (sd = 11.0) respectively. Mean satisfaction was 80.7 (sd = 21.9). Those who achieved II in pain, function and high demand activities by 3 months had satisfaction scores 87.8−89.3 compared to lower scores for all others; range 64.7−69.8 for those who did not achieve II. Fifty-seven, 47, 57 and 15% expected improvement in pain, other symptoms, mobility/ADL and social roles/IADL respectively. In model 1 (Fig. 1), earlier II achievement in pain (estimate −0.191) and function (estimate −0.389) was directly associated with higher satisfaction and higher expectation for mobility/ADL improvements (−0.106) was directly associated with lower satisfaction. Earlier achievement of II in function was associated with higher demand II achievement (0.628) but II in higher demand activities was not associated with satisfaction. All other significant effects were indirect. Model 2 showed that achievement of II in each of pain, function and high demand activities by 3 months was significantly associated with higher satisfaction as compared to achievement by 6 or 12 months (2–5 fold impact) or not achieving II (12–17 fold impact). Conclusions: People who achieve important improvement in pain, and function by 3 months post TKR are more satisfied with their outcome. Efforts to minimize recovery time should be tested to determine if more people recover more quickly and are satisfied with their outcomes.
Read full abstract