Skip to main content

Table 4 Posterior estimates for the ICU data based on joint models

From: Joint modeling of multivariate longitudinal data and the dropout process in a competing risk setting: application to ICU data

Modeling of dropouts

Cause specific hazards

Subdistribution hazards

Sub-model

    

LONGITUDINAL

Posterior mean

(95% CrI ))

Posterior mean

(95% CrI ))

Intercept

6.29

(6.05, 6.53)*

6.24

(5.74, 6.75)*

Time

-0.05

(-0.07, -0.04)*

-0.06

(-0.07, -0.05)*

Treatment group

0.34

(0.10, 0.58)*

0.26

(-0.17, 0.68)

Time × Treatment group

-0.012

(-0.04, 0.01)

-0.02

(-0.03, -0.003)*

Age (years)

0.45

(0.27, 0.63)*

0.53

(0.12, 0.96)*

Male Gender

0.44

(0.25, 0.63)*

0.47

(0.02, 0.96)*

Septic patients

1.33

(1.15, 1.51)*

1.44

(1.01, 1.86)*

SURVIVAL

HR

(95% CrI))

SHR

(95% CrI))

DEATHS

    

Treatment group

1.22

(0.98, 1.53)

1.23

(1.00, 1.53)*

Age (years)

1.93

(1.55, 2.41)*

1.38

(1.10, 1.74)*

Male Gender

1.19

(0.96, 1.50)

1.35

(1.07, 1.69)*

Septic patients

1.38

(1.11, 1.72)*

1.18

(0.95, 1.46)

γ(1)

3.32

(3.09, 3.57)

3.42

(3.24, 3.61)

DISCHARGES

    

Treatment group

1.17

(1.02, 1.36)*

1.25

(1.08, 1.43)*

Age (years)

0.91

(0.79, 1.05)

1.08

(0.93, 1.24)

Male Gender

1.25

(1.07, 1.45)*

1.30

(1.11, 1.51)*

Septic patients

1.02

(0.88, 1.18)

1.12

(0.96, 1.29)

γ(2)

-2.96

(-3.11, -2.80)

-3.35

(3.48, -3.22)

  1. The upper table displays longitudinal joint estimates, and the lower table displays the survival estimates in cause-specific hazards and sub-distribution hazards. * indicates statistically significance at level 0.05 (in the Bayesian sense; 95% credible set excludes 0)
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us