BACKGROUND Late nephrologist referral may adversely affect outcome in patients initiating maintenance hemodialysis therapy, mostly with temporary catheters that may further increase morbidity and mortality. Our aim was to evaluate the influence of 2 variables on mortality: presentation mode (planned versus unplanned) and type of access (arteriovenous fistula [AVF] versus temporary catheter) at entry. METHODS This was a 3-center, 5-year, prospective, observational, cohort study of 538 incident patients. Measurements included presentation mode, type of access, renal function and biochemical test results at entry, and stratification of risk groups. Main outcome measures were mortality and hospitalization. RESULTS Of 281 planned patients (52%), 73% initiated therapy with an AVF. Of 257 unplanned patients (48%), 70% initiated therapy with a catheter (P < 0.001). Multivariate Cox analysis showed that unplanned presentation (hazard ratio [HR], 1.73; 95% confidence interval [CI], 1.23 to 2.44) and initiation of therapy with catheter (HR, 1.75; 95% CI, 1.25 to 2.46) were independently associated with greater mortality and similar HRs after adjusting for confounders. At 12 months, the number of deaths was 3 times higher in both the unplanned versus planned groups and catheter versus AVF groups. The joint effect of unplanned dialysis initiation and catheter use had an additive impact on mortality (HR, 2.89; 95% CI, 1.97 to 4.22). Greater hematocrit (HR, 1.04; 95% CI, 1.01 to 1.09) and albumin level (HR, 1.79; 95% CI, 1.37 to 2.33) showed an independent association with survival, underscoring the benefits of predialysis care. Using Poisson regression, all-cause hospitalization (incidence rate ratio, 1.56; 95% CI, 1.36 to 1.79; P < 0.001) and infection-related (incidence rate ratio, 2.62; 95% CI, 1.91 to 3.59; P < 0.001) and vascular access-related (incidence rate ratio, 1.49; 95% CI, 1.15 to 1.94; P < 0.003) admissions were higher in unplanned patients initiating therapy with a catheter than in planned patients initiating therapy with an AVF, after adjusting for confounders. CONCLUSION Unplanned dialysis initiation and temporary catheter were independently associated with greater mortality rates in incident patients. The combined influence of both variables was associated with greater morbidity and mortality than either variable alone.