Skip to main content
. 2014 Jan 3;10(2):96–104. doi: 10.1080/17290376.2013.870119

Table 3.

Predictors of the HIV-related symptom burden.

Variables β coefficient (95% CI) p Adjusted β coefficient (95% CI)a p
Sociodemographic variables
Age 0.01 (–0.004, 0.014) 0.307
Male vs. female 0.05 (–0.14, 0.25) 0.589
Educational level 0.05 (0.01, 0.09) 0.014 0.12 (–0.01,0.24) 0.05
Never married/widowed/separated/divorced vs. married or cohabiting
0.05 (–0.14, –0.25) 0.602
Urban vs. rural residence −0.77 (–0.95, –0.59) <0.001 0.24 (–0.33, 0.81) 0.406
Formal/informal employment vs. not −1.03 (–1.23, –0.84) <0.001 −2.36 (–3.00, –1.72) <0.001
Clinical variables
Time since HIV diagnosis −0.09 (–0.21, 0.03) 0.147
CD4 count (cells/µL) −3.41 (–5.44, –1.39) 0.001 −0.01 (–0.011, –0.017) <0.001
Social variables
Internalised stigma scores 0.54 (0.49, 0.59) <0.001 0.66 (0.57, 0.76) <0.001
Discrimination experiences score −0.31 (–0.44, –0.17) <0.001 0.25 (–0.14, 0.64) 0.204
Alcohol use score 0.24 (0.07, 0.42) 0.007 0.33 (0.14,0.51) <0.001

Note: All variables with p < .05 in the baseline HIV-related symptom burden adjusted model were eligible for the multivariate model.

a

Goodness of fit quasi-likelihood under independence model criterion value = 44,318.46.