Skip to main content

Table 3 Event process estimates from the joint model and time-varying Cox model estimates

From: Joint modelling of longitudinal and time-to-event data: an illustration using CD4 count and mortality in a cohort of patients initiated on antiretroviral therapy

 β estimateaHR (95% CI)S.E.p-value
Time-varying proportional hazards model
Longitudinal \(\sqrt {CD4}\) count (per unit decrease)0.161.17 (1.12-1.23)0.03<0.001
Event Process
Age (years)0.011.01 (0.99-1.03)0.010.148
Log10 viral load (copies/ml)0.371.45 (1.21-1.69)0.120.002
Men (ref: women)0.051.05 (0.70-1.40)0.180.775
Urban site (ref: rural)-0.080.92 (0.49-1.35)0.220.758
Prevalent TB (ref: no prevalent TB)-0.210.81 (0.34-1.28)0.240.377
Association (per unit decease)0.301.34 (1.27-1.42)0.03<0.001
  1. aHR: adjusted hazard ratios; S.E.: standard error