OBJECTIVES:  To assess comparative effectiveness of peritoneal dialysis (PD) and hemodialysis (HD) in end-stage renal disease (ESRD) patients initiating dialysis therapy. METHODS:  We constructed retrospective cohort with national administrative healthcare databases. The study population comprised patients with ESRD in Korea who newly treated kidney dialysis such as hemodialysis and peritoneal dialysis between January 1, 2004 and December 31, 2015. We excluded patients who received kidney transplant; lost follow-up; death; stop dialysis within three months from the initiation date. The primary outcome was all cause mortality, and the secondary outcome was incidence of major adverse cardiac and cerebrovascular events (MACCE). We analyzed that the comparative risks of primary and secondary outcomes were determined with the use of Cox regression. Propensity-score matching was used to assemble patient cohorts with similar baseline characteristics. RESULTS:  A total of 96,626 eligible patients with ESRD were composed of 18,216 treated with PD, 78,410 treated with HD. The crude morality rates were 95.0 per 1000 patient-years (PY) in PD and 96.5/1000PY in HD. Compared with HD, PD was associated with increased risks of mortality (adjusted hazard ratio: 1.27, 95% CI: 1.24-1.31, p-value <0.0001) and MACCE rate (CVD, HR: 1.11, 95% CI: 1.06-1.16, p-value <0.0001). The results were similar in analysis of matching patients. However, in subgroup analysis, there was no statistical significant difference between HD and PD, especially in non-diabetic pa­tients less than age 65. CONCLUSIONS:  The results from this study show that patient treated with PD was associated with a high risk of mortality and MACCE. However, there was no statistical significant difference in group of non-diabetic pa­tients less than age 65. Although we could not assess cost effectiveness, considering cost and time to spend dialysis, appropriate selection of dialysis modality according to baseline characteristics is needed.