Table 8.
Differences in EQ-5D-5L utility and EQ-VAS scores between the general population and Roma respondents
| Roma | Lower 95% CI | Upper 95% CI | |
|---|---|---|---|
| Unadjusted differences (full sample: N = 2308) | |||
| EQ-5D-5L utility |
− 0.043*** (0.006) |
− 0.032 | − 0.055 |
| EQ-VAS score |
− 7.978*** (0.791) |
− 9.529 | − 6.427 |
| Regression modelling | |||
|---|---|---|---|
| Model | Roma coefficient | Lower 95% CI | Upper 95% CI |
| EQ-5D-5L utility (matched sample: N = 1612) | |||
| OLS |
− 0.046*** (0.009) |
− 0.065 | − 0.028 |
| GLM |
− 0.044*** (0.009) |
− 0.062 | − 0.026 |
| Tobit |
− 0.054*** (0.017) |
− 0.098 | − 0.030 |
| EQ-VAS score (matched sample: N = 1612) | |||
| OLS |
− 8.001*** (1.382) |
− 10.712 | − 5.289 |
| GLM |
− 7.629*** (1.428) |
− 10.427 | − 4.830 |
For regression models, coefficients are marginal effects from Ordinary Least Squares Models, Generalized Linear Models with a Gaussian Distribution and a Log Link Function and a Tobit Model right censored at 1. All models include CEM weights and are additionally adjusted for Age. Robust standard errors in parentheses
***p < 0.01