Background: Extracorporeal membrane oxygenation (ECMO), or extracorporeal life support (ECLS), has been gradually incorporated into cardiopulmonary resuscitation procedures, post-operational life support, and a variety of scenarios. Despite of the wide use of ECMO in Taiwan, the efficacy of ECLS in these cardiac and non-cardiac conditions was largely unknown. Therefore, the objective of this study was to compare the survival outcomes in the patients who received ECMO under different indications. Methods: The patients aged 20 years and above who received ECMO from September 1, 2009 to December 31, 2012 were identified from Taiwan’s National Health Insurance Database. The indications were classified into 1) life support after cardiac surgery (LSCS), 2) myocardial infarction and cardiogenic shock (post-MI), 3) injury, 4) infection, adult respiratory distress syndrome, and septic shock, and 5) others. Rare conditions such as donor support and obstetric labor complications were excluded. Cox-regression model was used to compare the 30-day and one-year mortality, using the LSCS as the reference group. Results: A total of 3,346 patients were included in the analysis (1,428, 944, 216, 691, and 67 for groups 1 to 5, respectively). The mean age was 53.2 years, and 71.2% was male. The 30-day mortality was 51.6%, 66.6%, 55.6%, 49.2%, and 49.3%, and the one-year mortality rate was 67.4%, 77.8%, 66.2%, 74.0%, and 68.7%, for the five groups, respectively. In the first 30 days, the post-MI group had a higher mortality hazard (HR: 1.62, 95% CI: 1.46~1.81) and the infection group had a lower one (HR: 0.87, 95% CI: 0.76~0.99) than the LSCS group. The one-year mortality was still the highest among the post-MI group (HR: 1.52, 95% CI: 1.38~1.67), and the survival among the other groups did not statistically differ from that of the LSCS group. Conclusion: The overall survival was worst in the patients who required ECLS due to myocardial infarction and cardiogenic shock. Patients with infection, adult respiratory distress syndrome, and septic shock had a higher possibility of surviving through the first month, but the overall mortality was not necessarily better. Further analysis regarding the cost and benefit of ECLS is warranted.