The high mortality rate of patients with end stage renal failure (ESRF) treated by dialysis is determined principally by irreversible factors such as age and comorbidity. In this single centre retrospective study of all 1260 ESRF patients who started dialysis between 1980 and 1999 it has been demonstrated that a short duration of specialist predialysis follow up is associated with a worse long term outcome on dialysis. Kaplan-Meier survival curves were plotted according to duration of predialysis follow up (group A, < or = 90 days; group B > 90 days), censoring for first transplant, and compared using a log rank test. Differences between groups were examined using an unpaired t test. Cox regression analysis was performed to examine the influence of selected variables on survival. Group A had the worst mortality (survival proportions of 87%, 74%, and 31% in A and 94%, 87%, and 55% in B at four months, one year, and five years respectively, p < 0.001). The increased risk of death was seen principally during the first few months of dialysis. ESRF associated with systemic disease was more prevalent in A. There were small but significant differences in predialysis clinical data, including age and serum albumin (p < 0.001). Fewer patients in A were suitable for transplant listing (p < 0.01). In the regression analysis, age, diabetes, predialysis serum albumin, suitability for transplant work-up and listing ("transplantability"), and the interval between referral and dialysis were significant predictors of survival. In summary, this study strengthens the previously reported association between late referral of ESRF patients and subsequent poor survival on dialysis. This important message is relevant to all potential referring physicians.