Skip to main content

Table 2 Variable definitions in the conditional logit regression model

From: Reconsidering the use of rankings in the valuation of health states: a model for estimating cardinal values from ordinal data

Variable Definition
M2 1 if mobility is at level 2 or higher; 0 otherwise
S2 1 if self-care is at level 2 or higher; 0 otherwise
U2 1 if usual activities is at level 2 or higher; 0 otherwise
P2 1 if pain/discomfort is at level 2 or higher; 0 otherwise
A2 1 if anxiety/depression is at level 2 or higher; 0 otherwise
M3 1 if mobility is at level 3; 0 otherwise
S3 1 if self-care is at level 3; 0 otherwise
U3 1 if usual activities is at level 3; 0 otherwise
P3 1 if pain/discomfort is at level 3; 0 otherwise
A3 1 if anxiety/depression is at level 3; 0 otherwise
N3 1 if any domain is at level 3; 0 otherwise
  1. Note: This specification is algebraically equivalent, but not identical, to that used by Dolan [30] in modeling time trade-off values. The slight modification from Dolan is adopted so that the modeled value for the EQ-5D state 33333 is simply equal to the sum of all of the estimated coefficients from the model, which allows for convenient rescaling (see text).