Determinants of CD4 cell count change and time-to default from HAART; a comparison of separate and joint models

BMC Infectious Diseases

Table 5 Posterior effects of the predictor variables from modeling CD4 cell count change with models (2) (no patient specific variance) and (5) (with patient specific variance)

Parameters	Estimate	Standard error	P-values
Intercept	1.808	0.910	0.470
Age	− 0.658	0.145	0.831
Follow up times	0.047	0.052	0.058
Previous CD4 cell count (ref=y_j ≤ y_j − 1) or I_i = 0
I_i = 1 (y_j > y_j − 1)	0.072	0.320	0.650
Gender (ref = female)
Male	0.003	0.084	0.976
Residence area (ref = Urban)
Rural	−0.380	0.083	0.648
Marital status(ref = living without partner)
Living with partner	0.109	0.086	0.208

where y_j is the CD4 cell count at the j^th follow-up visit; y_j − 1 is the CD4 cell count at the (j-1)^th follow-up visit

ISSN: 1471-2334