Table 3.
Modeling self-reported EQ-VAS scores by country and EQ-5D version
| Japan EQ-5D-3L | UK EQ-5D-3L | Japan EQ-5D-5L | England EQ-5D-5L | Spain EQ-5D-5L | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Coef | P value | Coef | P value | Coef | P value | Coef | P value | Coef | P value | |
| Mobility | −0.88 | 0.785 | −5.87 | 0.000 | −2.87 | 0.029 | −3.46 | 0.000 | −0.35 | 0.719 |
| Self-care | −5.42 | 0.394 | −6.82 | 0.000 | −4.41 | 0.128 | −0.21 | 0.836 | −3.29 | 0.065 |
| Usual activities | −9.50 | 0.018 | −8.54 | 0.000 | −9.61 | 0.000 | −4.87 | 0.000 | −5.86 | 0.000 |
| Pain/discomfort | −11.03 | 0.000 | −6.80 | 0.000 | −4.53 | 0.000 | −3.28 | 0.000 | −4.59 | 0.000 |
| Anxiety/depression | −7.22 | 0.003 | −8.02 | 0.000 | −5.83 | 0.000 | −6.96 | 0.000 | −7.87 | 0.000 |
| Constant | 115.80 | 0.000 | 125.80 | 0.000 | 114.94 | 0.000 | 105.99 | 0.000 | 109.82 | 0.000 |
| Adjusted R-squareda | 0.2027 | 0.4226 | 0.2285 | 0.5031 | 0.4026 | |||||
| N | 543 | 3372 | 1026 | 996 | 1000 | |||||
Note: (1) Shapiro–Wilk test rejected the null hypothesis of normally distributed residuals in all five regressions at the 5% significance level. (2) The Breusch–Pagan test accepted the null hypothesis that the variance of the residuals is homogeneous at the 5% significance level for the Japan EQ-5D-3L model, but rejected the null hypothesis in all other four models. (3) The regression equation specification error test (RESET) rejected the null hypothesis of linear functional form for the specifications in Japan EQ-5D-5L model and Spain EQ-5D-5L model at the 5% significance level. (4) None of the five models indicated issues with multicollinearity (mean variance inflation factor < 3)
aAdjusted R-squared is computed using the formula 1 − [(1 − Rsq) × (N − 1)/(N − k − 1)], where Rsq represents R-squared (the proportion of variance in the dependent variable that can be explained by the independent variables); N represents the number of observations; and k represents the number of independent variables