Skip to main content

Table 7 A summary of the results obtained from fitting the different statistical models to the binary outcome data from NOSH CRCT(N = 9207)

From: Analysing cluster randomised controlled trials using GLMM, GEE1, GEE2, and QIF: results from four case studies

 

Unadjusted model

Adjusted modela

Parameters

GLMM

GEE1

GEE2

QIF

GLMM

GEE1

GEE2

QIF

Intervention effect2

        

1.37

1.36

1.36

1.36

1.31

1.31

1.27

1.28

SE

        

0.08

0.08

0.08

0.08

0.07

0.07

0.05

0.07

P-value

        

0.0002

 < 0.0001

 < 0.0001

0.0009

0.0002

 < 0.0001

 < 0.0001

0.0002

95% CI

        

1.16 to 1.60

1.17 to 1.58

1.17 to 1.58

1.17 to 1.59

1.14 to 1.51

1.14 to 1.49

1.15 to 1.41

1.12 to 1.46

ICC

        

0.0262

0.0192

0.0383

0.0192

0.0162

0.0098

0.0042

0.0098

Number of subjects

        

9207

9207

9207

9207

9207

9207

9207

9207

Number of clusters

        

92

92

92

92

92

92

92

92

  1. aThe statistical models were adjusted for the cluster-level baseline breastfeeding rate and local government area. Note that SE Standard error, CI Confidence interval, ICC Intracluster correlation coefficient, GLMM Generalized linear mixed model, mGLM Marginal generalized linear model, GEE Generalized estimating equations, QIF Quadratic inference function
  2. 1. The binary outcome was if the mother was breastfeeding her baby at 6 weeks postnatally (response value = 1) or not (response value = 0)
  3. 2. The intervention effect for the binary outcome is the odds for breastfeeding at 6 weeks postnatally in the NOSH intervention group compared to the odds of breastfeeding in the control group with an odds ratio > 1 favouring better outcomes (higher odds of breastfeeding) in the intervention group
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us