BACKGROUND: Pooled analysis of phase 3, double-blind, double-dummy ECHO and THRIVE trials comparing rilpivirine (TMC278) and efavirenz. METHODS: Treatment-naive HIV-1-infected adults were randomized 1:1 to rilpivirine 25 mg once daily or efavirenz 600 mg once daily, with background tenofovir disoproxil fumarate/emtricitabine (TDF/FTC) (ECHO) or TDF/FTC, zidovudine/lamivudine, or abacavir/lamivudine (THRIVE). The primary endpoint was confirmed response [viral load <50 copies per milliliter; intent-to-treat time-to-loss-of-virologic-response (ITT-TLOVR) algorithm] at week 48. The pooled data set enabled analyses of subgroups and predictors of response/virologic failure. RESULTS: Confirmed responses were 84% (rilpivirine) and 82% (efavirenz). The difference in response rates (95% confidence interval) was 2.0% (-2.0% to 6.0%). The incidence of virologic failure was 9% (rilpivirine) versus 5% (efavirenz). Responses in ITT-TLOVR and ITT-snapshot analyses were consistent. Responses were similar for rilpivirine and efavirenz by background regimen, gender, race and clade. Suboptimal adherence and higher baseline viral load resulted in lower responses, higher virologic failure, and development of resistance in both groups; the effects on virologic failure were more apparent with rilpivirine. CD4 cell count increased over time in both groups. Rilpivirine compared with efavirenz gave smaller incidences of adverse events leading to discontinuation (3% vs. 8%, respectively), treatment-related grade 2-4 adverse events (16% vs. 31%), rash (3% vs. 14%), dizziness (8% vs. 26%), abnormal dreams/nightmares (8% vs. 13%), and grade 2-4 lipid abnormalities. CONCLUSIONS: At week 48, rilpivirine 25 mg once daily and efavirenz 600 mg once daily had comparable response rates. Rilpivirine had more virologic failures and improved tolerability versus efavirenz.