Table 2.

Rate of Achieving CD4⁺ T-Cell Count Normalization

Model	Rate Ratio (95% CI)^a	P Value
Univariate
Sex, female vs male	1.18 (0.73–1.89)	.51
Ethnicity
African American vs European American	0.92 (0.73–1.15)	.44
Other vs European American	0.80 (0.56–1.14)	.22
Age at ART initiation, each increase of 1 y	1.00 (0.99–1.02)	.86
Time from EDS to ART initiation, ≤12 vs>12 mo	1.53 (1.22–1.91)	<.001
Time from entry to ART initiation, ≤12 vs >12 mo	1.51 (1.21–1.89)	<.001
Study entry CD4⁺, ≥500 vs <500 cells/μL	2.45 (1.98–3.05)	<.001
Pre-ART CD4⁺, each increase of 10 cells	1.06 (1.05–1.06)	<.001
Pre-ART CD4⁺, ≥500 vs <500 cells/μL	5.91 (4.77–7.32)	<.001
Pre-ART viral load, each increase of 1 log₁₀ copies/mL	1.05 (0.90–1.21)	.57
Calendar year of ART initiation, each increase of 1 y	1.06 (1.04–1.09)	<.001
Antiretroviral regimen^b
PI-based vs NNRTI-based	1.39 (1.06–1.82)	.02
Other vs NNRTI-based	0.61 (0.47–0.80)	<.001
Other vs PI-based	0.44 (0.34–0.57)	<.001
Duration of VL-suppressive ART, each increase of 1 y^c	0.88 (0.84–0.92)	<.001
Time from ART initiation to VL suppression, each increase of 1 mo	0.98 (0.97–0.98)	<.001
Multivariate^d
Model1: time from EDS to ART initiation, ≤12 vs>12 mo	1.32 (1.04–1.67)	.02
Model 2: time from study entry to ART initiation, ≤12 vs >12 mo	1.77 (1.38–2.26)	<.001
Model 3: study entry CD4⁺, ≥500 vs <500 cells/μL	2.00 (1.51–2.64)	<.001
Model 4: pre-ART CD4⁺, ≥500 vs <500 cells/μL	4.08 (3.14–5.30)	<.001

Abbreviations: ART, antiretroviral therapy; EDS, estimated date of seroconversion; NNRTI, nonnucleoside reverse transcriptase inhibitor; PI, protease inhibitor; VL, viral load.

The Cox proportional hazards model was used to estimate the rate ratio for achievement of the first CD4⁺ count of 900 cells/μL or more after ART initiation.

PI-based was defined as PI therapy without any NNRTI, NNRTI-based was defined as NNRTI therapy without any PI, and the other treatment options were a combination of PI and NNRTI therapy or neither of these drug classes.

Duration of VL-suppressive ART was indexed from the date of starting ART.

The covariates adjusted for in models 1 and 2 were study entry CD4⁺ counts 500 cells/μL or more or less than 500 cells/μL, pre-ART CD4⁺ count 500 cells/μL or more or less than 500 cells/μL, calendar year of ART initiation, ART regimen, length of follow-up (years), and duration of VL-suppressive ART. The covariates adjusted for in model 3 were pre-ART CD4⁺ counts 500 cells/μL or more or less than 500 cells/μL, and time from study entry to ART initiation within 12 months or more than 12 months, calendar year of ART initiation, ART regimen, length of follow-up (years), and duration of VL-suppressive ART. Similar results were found when the model was adjusted for time from EDS to ART initiation within 12 months or more than 12 months instead of time from entry to ART. The covariates adjusted for in model 4 were study entry CD4⁺ counts 500 cells/μL or more or less than 500 cells/μL; time from study entry to ART initiation within 12 months or more than 12 months, calendar year of ART initiation, ART regimen, length of follow-up (years), and duration of VL-suppressive ART. Similar results were found when the model was adjusted for time from EDS to ART initiation within 12 months or more than 12 months instead of time from entry to ART.