Skip to main content

Table 7 Results of random parameter logit model subgroup analysis for duration of MDR-TB treatment

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

Attributes and levelsa

On treatment < 6 months

On treatment ≥ 6 months

P value

 

Coefficient

SD

Relative importance

Coefficient

SD

Relative importance

 

DOT provider

  

16.7

  

36.7

 

 Family (reference)

–

–

 

–

–

  

 Community health work

0.58

1.94

 

1.49

1.77

 

0.018

 Expert client

0.32

1.79

 

1.16

1.87

 

0.017

Location

  

48.0

  

27.8

 

Workplace (reference)

 Home

1.67

1.27

 

1.13

1.16

 

0.169

Support

  

35.3

  

35.5

 

SMS (reference)

 Phone call reminders

0.73

0.99

 

0.66

0.11

 

0.546

 Travel vouchers

1.23

0.33

 

1.44

0.62

 

0.543

  1. aDummy coded attributes (coefficient of the reference category is constrained to be 0)
  2. Coefficients – indicates preference weights of attribute levels.
  3. CI Confidence interval, SD standard deviation for preference heterogeneity (random component of the model coefficients)
  4. Number of observations = 744 (on treatment< 6 months), 1704 (on treatment ≥6 months). Total observation <overall (2472) because 1 participant had missing duration on treatment
Back to article page

BMC Health Services Research

ISSN: 1472-6963

Contact us