Comparison of the effectiveness of non-nucleoside reverse transcriptase inhibitor-containing and protease inhibitor-containing regimens using observational databases.

摘要: Objectives To compare the effectiveness of first protease inhibitor (PI)-containing and non-nucleoside reverse transcriptase (NNRTI)-containing regimens. Methods Data were analysed from three large HIV patient databases: Apache Insight (APACHE), Target Management Services (TMS) Clinical Partners (CP). The therapy was time taken for HIV-1 RNA to fall below detectable levels on highly active antiretroviral regimen (PI- or NNRTI-containing) subsequent failure (two consecutive measurements). Comparisons made using proportional hazards models, adjusting differences in age, sex, previous use, calendar year baseline viral load CD4 T-cell count. Results type not associated with undetectable any databases, all which had high power detect a difference. PI-containing regimens significantly less likely fail after reaching loads APACHE CP patients (relative hazard, 1.7; 95% confidence interval, 1.3--2.1 relative 1.6; 1.0--2.5 respectively). These results remained significant allowing an unmeasured confounder moderate effect risk. No association between found TMS patients, possibly due low (67% hazard 1.5). difference increase > 100 x 10(9)cells/l In database, those NNRTI-containing more have failing response. Conclusions lower risk treatment than