Skip to main content

Table 2 Results for the different types of joint models for NTB memory domain

From: Using joint models to disentangle intervention effect types and baseline confounding: an application within an intervention study in prodromal Alzheimer’s disease with Fortasyn Connect

 a) Current valueb) Current value plus slopec) Cumulative effect
 Coefficient (SE)p ValueCoefficient (SE)p ValueCoefficient (SE)p Value
Longitudinal sub-model:      
Time β1-0.101 (0.030)0.001-0.128 (0.030)0.000-0.092 (0.029)0.002
Fortasyn β20.042 (0.082)0.6170.039 (0.083)0.6400.042 (0.083)0.610
Time × Fortasyn β30.052 (0.043)0.2190.049 (0.043)0.2540.054 (0.042)0.192
Bmmse β40.160 (0.021)0.0000.159 (0.021)0.0000.158 (0.021)0.000
 Log Hazard (SE)p ValueLog Hazard (SE)p ValueLog Hazard (SE)p Value
Survival sub-model:      
Fortasyn γ10.154 (0.203)0.4490.513 (0.387)0.1850.112 (0.198)0.573
Ass. α1-1.214 (0.174)0.000-1.162 (0.341)0.001--
Ass. slope α2---6.792 (1.854)0.000--
Ass. AUC α3-----0.671 (0.118)0.000
Bmmse γ2-0.098 (0.057)0.085-0.156 (0.088)0.078-0.163 (0.055)0.003
\