Skip to main content

Table 6 Error measurements for predicting utility values based on OHS and OKS, according to the distribution of expressed utilities. Validation sample

From: Mapping analysis to predict EQ-5D-5 L utility values based on the Oxford Hip Score (OHS) and Oxford Knee Score (OKS) questionnaires in the Spanish population suffering from lower limb osteoarthritis

 

Hip osteoarthritis, OHS,n = 301

MAE

Utility score ≥ mediana

Utility score < mediana

Model

Dependent Variable

  

OLS

Utility

0.1137 (0.0981–0.1293)

0.1576 (0.1374–0.1777)

Tobit

Utility

0.1031 (0.0885–0.1177)

0.1518 (0.1325–0.1711)

GLM

Utility

0.0857 (0.0742–0.0973)

0.1410 (0.1213–0.1711)

Beta reg

Utility

0.0941 (0.0801–0.1080)

0.1536 (0.1330–0.1741)

 

Knee osteoarthritis, OKS,n = 313

MAE

Utility score ≥ medianb

Utility score < medianb

Model

Dependent Variable

  

OLS

Utility

0.1106 (0.0969–0.1243)

0.1470 (0.1279–0.1661)

Tobit

Utility

0.1031 (0.0899–0.1164)

0.1451 (0.1259–0.1642)

GLM

Utility

0.0869 (0.0758–0.0980)

0.1389 (0.1204–0.1575)

Beta reg

Utility

0.0899 (0.0782–0.1016)

0.1389 (0.1208–0.1570)

  1. aMedian utility value (Hip osteoarthritis): 0.6973
  2. bMedian utility value (Knee osteoarthritis): 0.6852
  3. OLS Ordinary least squares
  4. GLM Generalized linear model. Link function: log. Distributional family: Gaussian
  5. Beta reg Beta regression
  6. MAE Mean absolute error
  7. CI 95% in parentheses