The optimal time of dialysis initiation among patients with chronic kidney disease (CKD) is unclear in recent years. We performed a meta-analysis to assess the association of early vs. late initiation of dialysis with estimated glomerular filtration rate. PUBMED, EMBASE, the Cochrane Library, and article reference lists were searched for relevant observational trials. A pooled hazard ratio (HR) with 95% CI was used to estimate the mortality risk. Twenty-six cohort studies and one randomized controlled trial were identified. Early start of dialysis was associated with the increased risk of mortality (HR = 1.23, 95% CI: 1.04–1.43) compared with late start of dialysis. In the subgroup analysis, age younger than 65 years at the early start of dialysis demonstrated higher mortality (HR = 1.20, 95% CI: 1.05–1.35) than the late start. Compared with peritoneal dialysis, the pooled HR with HD was 1.25 (95% CI: 1.17–1.34). Early start of dialysis increased the mortality risk compared with late start among patients with CKD.