Abstract(#br)Background(#br)China's National Free Antiretroviral Treatment Program has substantially scaled up and standardised antiretroviral treatment (ART) since 2008. Meanwhile, no study worldwide has examined the effects of rapid ART programme scale-up on treatment outcomes in resource-limited settings on a large scale.(#br)Methods(#br)The national HIV drug resistance surveillance database included annual cross-sectional surveys HIV drug resistance from 2003–12. We used China's national HIV drug resistance surveillance database to determine virological failure, acquired drug resistance, and poor adherence after 12–15 months of first-line ART. The probability proportional to proxy size (PPPS) sampling recommended by WHO was used in this study. χ tests and logistic regression were used... for categorical variables. The institutional review board of the National Center on HIV/STD Control and Prevention (NCAIDS), China CDC, approved this study.(#br)Findings(#br)Of the 2252 patients included in this study, 1180 (52%) were aged less than 40 years; 2100 (93%) belonged to the Han ethnic group; 1331 (59%) were male; 963 (43%) had primary school (1–6 years of education) or less; 1665 (74%) were married; 968 (43%) were farmers; 929 (41%), 947 (42%), and 322 (14%) were infected through blood or plasma transmission, sexual intercourse, and intravenous drug use, respectively. 1431 (64%) patients had initiated ART before 2008, and 821 (36%) initiated ART after 2008. The incidence of virological failure after 12–15 months of treatment was lower in the group of patients who had initiated ART after 2008 than in the group of patients who initiated ART before 2008 (12·1% vs 26·6%). The incidence of HIV drug resistance to ART was also significantly reduced in patients who initiated ART after 2008 compared to patients who intiated treatment before 2008 (5·4% vs 15·4%). However, these improvements are strongly associated with the standardised use of lamivudine-based regimens instead of didanosine-based regimens. Patients who initiated lamivudine-based regimens before 2008 had significant improvement in adherence (missed doses adjusted odds ratio [AOR] 0·65, 95% CI 0·45–0·96), virological failure (AOR 0·29 95% CI 0·22–0·39), and drug resistance outcomes (AOR 0·29 95% CI 0·20–0·42) compared with those who initiated didanosine-based regimens. We found no significant changes in these outcomes between patients who initiated lamivudine-based regimens before 2008 and those who initiated this treatment after 2008.(#br)Interpretation(#br)China's National Free Antiretroviral Treatment Program has been largely successful throughout the scale-up, with an overall reduction in virological failure and HIV drug resistance. However, excluding the effect of lamivudine-based regimens, the programme must improve patient adherence and quality of care, particularly in key vulnerable populations such as those infected through injecting drug use or blood routes. Further monitoring of treatment outcomes is needed to elucidate the determinants of long-term programmatic success. This is the first large scale study in the world evaluating the outcomes of ART treatment during a rapid programmatic scale-up.(#br)Funding(#br)Ministry of Science and Technology of China (grants 2012ZX10001-002 and 2009DFB30420), Chinese State Key Laboratory for Infectious Disease (development grant 2012SKLID103), and International Development Research Center of Canada (grant 104519-010). The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.