| Correlations (ρ)a | 0.5 SD | Anchor-based method | Predictive modelling |
---|
GRC1 | GRC2 | MIC | MIC | 95% CI | MIC | 95% CI |
---|
EQ-5D-5L | 0.53 | 0.49 | 0.108 | 0.118 | 0.038–0.198 | 0.103 | 0.062–0.237 |
EQ-5D-CW | 0.52 | 0.45 | 0.115 | 0.124 | 0.036–0.213 | 0.101 | 0.018–0.325 |
SF-6D | 0.45 | 0.51 | 0.048 | 0.064 | 0.029–0.099 | 0.056 | 0.020–0.143 |
GHCr-index | 0.36 | 0.31 | 8.876 | 9.158 | 1.798–16.517 | 8.786 | 3.096–17.804 |
EQ-VAS | 0.45 | 0.36 | 9.106 | 8.789 | 2.580–14.999 | 9.076 | 3.525–18.624 |
- a Correlations of GRC variables with score changes of utility and criteria variables (a value of ρ 0.30 and above is sufficiently large to use the anchors for the calculation of MIC). GRC Global rating of change (1 = evaluation of amalgam removal results, & 2 = item-2 of the SF-36); CI (bootstrapped) confidence interval (with 1000 iterations); SD (baseline) standard deviation; MIC minimally important change; EQ-5D-5L EuroQol 5-dimensional 5-level questionnaire; SF-6D Short-form 6-dimension; GHCr (reversed) general health complaints; EQ-VAS (EuroQol) visual analogue scale