Skip to main content

Table 3 Model performance: VitiQoL mapping algorithms

From: Estimating health related quality of life effects in vitiligo. Mapping EQ-5D-5 L utilities from vitiligo specific scales: VNS, VitiQoL and re-pigmentation measures using data from the HI-Light trial

  

Van Hout Crosswalk

Alava Crosswalk

  

M1 (Linear)

M2 (Multivariate)

M3 (Bayesian Linear)

M1 (Linear)

M2 (Multivariate)

M3 (Bayesian Linear)

VitiQoL

MAE

0.102

0.107

0.101

0.022

0.093

0.096

 

RMSE

0.1507

0.146

0.1506

0.149

0.143

0.149

 

Predicted Mean (SE)

0.8953 (0.002)

0.8939 (0.002)

0.8951 (0.002)

0.8916 (0.002)

0.8913 (0.002)

0.8912 (0.002)

 

Observed Mean (SE)

0.9067 (0.005)

0.9067 (0.005)

0.9067 (0.005)

0.9008 (0.005)

0.9008 (0.005)

0.9008 (0.005)

 

Difference (Mean)

0.0114

0.0128

0.0116

0.0092

0.0095

0.0096

 

AIC/DIC#

-1361

-1257

-1377#

-1382

-1319

-1399#

 

Predicted QALY

1.5728 (0.006)

1.5782 (0.007)

1.5725 (0.006)

1.5660 (0.005)

1.5754 (0.007)

1.5654 (0.006)

 

Observed QALY

1.5979 (0.017)

1.5979 (0.017)

1.5979 (0.017)

1.5887 (0.017)

1.5887 (0.017)

1.5887 (0.017)

 

Difference (QALY)

0.0251

0.0197

0.0254

0.0227

0.0133

0.0233

  1. M1 Linear Model, M2 Linear Multivariate Model, M3 Bayesian Linear Model, MAE Mean Absolute Error, RMSE Root Mean Squared Error, QALY Quality Adjusted Life Year, AIC Akaike Information Criterion, SD Standard Deviation, SE Standard Error. Difference (Mean): Observed - Predicted, QALY estimates derived from baseline, month 9 and month 21 data
Back to article page

Health and Quality of Life Outcomes

ISSN: 1477-7525

Contact us