Skip to main content

Table 4 Results of random parameter logit model (mixed logit model) for the MDR-TB CB-DOT model attributes

From: Developing a patient-centered community-based model for management of multi-drug resistant tuberculosis in Uganda: a discrete choice experiment

Attributes and levelsa

Coefficients

95%CI

P-value

SD

95%CI

P value†

Relative importance

Constant

0.52

0.30–0.74

< 0. 01

-

-

-

-

DOT provider

      

30.3%

Family (reference)

 Community health work

1.13

0.68–1.59

< 0. 01

1.85

1.33–2.37

< 0.01

 

 Expert client

0.89

0.46–1.32

< 0. 01

1.77

1.24–2.30

< 0.01

 

Location

      

33.5%

Workplace (reference)

 Home

1.25

0.91–1.59

< 0. 01

1.19

0.80–1.57

< 0.01

 

Support

      

36.2%

SMS (reference)

 Phone call reminders

0.70

0.41–0.99

< 0. 01

0.19

0.95–1.33

0.74

 

 Travel vouchers

1.35

1.04–1.65

< 0. 01

0.55

0.04–1.05

0.033

 
  1. aDummy coded attributes (coefficient of the reference category is constrained to be 0)
  2. CI Confidence interval, SD standard deviation for preference heterogeneity (random component of the model coefficients)
  3. † P value testing the hypothesis that standard deviation (heterogeneity across individuals’ preferences) equals ‘0’
  4. Number of observations = 2472