Table 3.
Results of the sequential multiple regression analysis (N = 85a): Variance explained in the global effectiveness of total hip arthroplasty at 12 months follow-up by sociodemographic and medical variables, preoperative expectations, change in symptoms and the fulfilment of expectations (multiple imputation of missing values)
| Variables included upon each step | Range of R2 across the 10 imputed data sets | Pooled β (p-value) in final model (only significant predictor variables shown) | |
|---|---|---|---|
| Step 1 Confounding factors |
Degree of school education Average hip pain in the last 3 months before surgery Overall severity of chronic pain (CPG) Hip function and mobility (WOMAC) Walking ability (Timed up and go score) |
0.261–0.304 | – |
| Step 2 Preoperative expectations |
Degree of school education Average hip pain in the last 3 months before surgery Overall severity of chronic pain (CPG) Hip function and mobility (WOMAC) Walking ability (Timed up and go score) Preoperative hip pain expectation |
0.372–0.416 |
β = −2.143 (p = 0.033) Walking ability (Timed up and go score) β = 3.745 (p < 0.001) Preoperative hip pain expectation |
| Step 3 Improvement in symptoms (Symptom change scores) |
Degree of school education Average hip pain in the last 3 months before surgery Overall severity of chronic pain (CPG) Hip function and mobility (WOMAC) Walking ability (Timed up and go score) Preoperative hip pain expectation Symptom change scores: Average hip pain in the last 3 months hip function and mobility (WOMAC) Health-related quality of life (SF-12 physical) |
0.419–0.493 |
β = − 2.338 (p = 0.020) Walking ability (Timed up and go score) β = 3.370 (p = 0.001) Preoperative hip pain expectation |
| Step 4 Fulfillment of expectations (calculated expectations-actuality discrepancy scores) |
Degree of school education Average hip pain in the last 3 months before surgery Overall severity of chronic pain (CPG) Hip function and mobility (WOMAC) Walking ability (Timed up and go score) Preoperative hip pain expectation Symptom change scores: Average hip pain in the last 3 months Hip function and mobility (WOMAC) Health-related quality of life (SF-12 physical) Calculated expectations-actuality discrepancy scores: Walking ability Independence Physical exercise General function Social interactions |
0.613–0.689 |
β = −3.103 (p = 0.002) Walking ability (Timed up and go score) β = 4.605 (p < 0.0001) Preoperative hip pain expectation β = 2.601 (p = 0.009) Fulfillment of expectations (calculated expectations-actuality discrepancy scores): Walking ability β = 2.952 (p = 0.002) Fulfillment of expectations (calculated expectations-actuality discrepancy scores):: Independence β = − 2.783 (p = 0.006) Fulfillment of expectations (calculated expectations-actuality discrepancy scores): General function |
| Final model (only significant predictors included) |
Walking ability (Timed up and go score) Preoperative hip pain expectation Calculated expectations-actuality discrepancy scores: Walking ability Independence General function |
0.510–0.544 |
aIndividuals with more than 3 out of 14 missing predictors were excluded from the analysis. The variables marked in bold are add up to the model at each step. Predictor variables individually significantly associated with global effectiveness of THA (see Supplementary Table 1 and 2) were entered in four steps. The significant predictors in the final model were: Hip pain expectation, walking ability at baseline and the calculated expectations-actuality discrepancy scores of walking ability, general function and independence in everyday life
β in final model = β regression coefficient after all listed variables have been entered; R2change = Increase in explained variance by step. Level of significance was set to p < 0.05