Skip to main content

Table 2 Regression model statistics with EQ-5D weights as the dependent variable

From: Mapping health outcome measures from a stroke registry to EQ-5D weights

Regression model OLS Tobit CLAD
  Coef. 95% CI Coef. 95% CI Coef. 95% CI
   Lower Upper   Lower Upper   Lower Upper
Constant 0.902 0.843 0.961 1.068 0.970 1.167 1.000 0.997 1.000
Toilet assistance −0.151 −0.364 0.062 −0.138 −0.325 0.048 −0.179 −0.418 0.061
Dressing assistance −0.187 −0.301 −0.072 −0.218 −0.362 −0.074 −0.247 −0.446 −0.048
Indoor mobility only −0.191 −0.266 −0.116 −0.230 −0.344 −0.117 −0.152 −0.240 −0.064
No mobility −0.363 −0.570 −0.156 −0.393 −0.580 −0.206 −0.281 −0.587 0.025
General health fairly good −0.055 −0.126 0.016 −0.129 −0.233 −0.024 −0.121 −0.229 −0.014
General health fairly bad −0.155 −0.263 −0.046 −0.233 −0.367 −0.099 −0.237 −0.367 −0.107
General health bad −0.246 −0.463 −0.028 −0.303 −0.529 −0.077 −0.370 −0.651 −0.088
Moody sometimes −0.074 −0.131 −0.017 −0.137 −0.215 −0.060 −0.107 −0.255 0.011
Moody often −0.208 −0.354 −0.062 −0.274 −0.413 −0.134 −0.223 −0.400 −0.045
Moody always −0.327 −0.503 −0.150 −0.387 −0.610 −0.163 −0.471 −0.788 −0.154
Proxy response −0.132 −0.271 0.007 −0.141 −0.280 −0.001 −0.032 −0.266 0.202
R2 / Pseudo R2 0.724    0.657    0.486   
  1. Note: OLS = ordinary least square; CLAD = censored least absolute deviation; Coef = coefficient; CI = confidence interval.