Skip to main content

Table 4 Predictive accuracy of the algorithms by total WOMAC score interval

From: Assessing the external validity of algorithms to estimate EQ-5D-3L from the WOMAC

Total WOMAC n   Barton Xie Wailoo
      WA CEC
0–10 195 ME (95 % CI) −0.081 (−0.106 to −0.057) −0.053 (−0.078 to −0.029) 0.048 (0.024 to 0.072) 0.093 (0.069 to 0.117)
MAE 0.143 0.132 0.110 0.117
RMSE 0.191 0.181 0.175 0.195
10–30 326 ME (95 % CI) −0.021 (−0.034 to −0.009) 0.013 (0.002 to 0.025) 0.004 (−0.009 to 0.017) 0.019 (0.003 to 0.034)
MAE 0.068 0.057 0.080 0.101
RMSE 0.113 0.109 0.117 0.141
30–50 336 ME (95 % CI) −0.085 (−0.104 to −0.066) 0.034 (0.016 to 0.053) −0.119 (−0.137 to −0.100) −0.014 (−0.033 to 0.005)
MAE 0.160 0.093 0.184 0.112
RMSE 0.195 0.175 0.210 0.178
50–100 211 ME (95 % CI) −0.143 (−0.175 to −0.111) 0.135 (0.100 to 0.170) −0.218 (−0.250 to −0.185) −0.307 (−0.350 to −0.264)
MAE 0.258 0.187 0.295 0.351
RMSE 0.277 0.294 0.322 0.443
  1. CI confidence interval, ME mean error, MAE mean absolute error, RMSE root mean square error, WA weighted average, CEC conditional on estimated component