Table 2.
Multivariate Logistic Regression Models of Virologic Response
| Variable | Lopinavir (n=70) | Atazanavir (n=154) | ||
|---|---|---|---|---|
| Adjusted odds ratios1 (±95%CI) | p-value | Adjusted odds ratios1 (±95%CI) | p-value | |
| Pre-treatment HIV RNA level ≥ 100,000 copies/ml vs <100,000 | 0.11 (0.013 –0.93) | 0.042 | 1.38 (0.37 – 5.11) | 0.63 |
| Adherence over past 6 months (self-report) ≥95% vs <95% | 2.9 (0.74 – 11.3) | 0.13 | 2.7 (1.04 – 6.9) | 0.042 |
| Pre-treatment CD4 count (cells/mm3) <200 vs ≥200 | 0.15 (0.033 – 0.66) | 0.012 | 0.36 (0.14 – 0.93) | 0.036 |
| Per tertile of hair level | ||||
| Lowest tertile | 1.00 | 1.00 | ||
| Middle tertile | 2.6 (0.59 – 11.9) | 0.21 | 2.7 (1.00 – 7.3) | 0.050 |
| Highest tertile | 39.8 (2.8 - 564) | 0.006 | 7.7 (2.0 – 29.7) | 0.003 |
1
Adjusted for age, race, exact number of months on therapy and degree of PI experience
*
Models using RTV levels from participants on LPV/RTV or ATV/RTV yield similar results