Background: While a low estimated glomerular filtration rate (eGFR) confers an increased risk of dying, the mortality associated with high eGFR values has not been ascertained. Methods: Four variable MDRD-derived eGFR was calculated in a total of 33,386 patients (18,620 F, 14,766 M) aged ≧50 years (median 68 years, IQR 60–76 years) collected by family doctors in 2000. They were then classified according to their eGFR into 30 ml/min/1.73 m2 bands (<30, 30–59, 60–89, 90–119, 120–150). The subsequent mortality status of each patient was determined at 31st December 2006 and cause of death recorded where available. Results: Applying Cox proportional hazards models (adjusted for age and gender), the hazard ratio (HR) of dying compared to a reference of patients with eGFRs of 60–89 ml/min/1.73m2 was, as expected, higher in the low eGFR bands (HR 1.37 (95% CI 1.29–1.45) for 30–59; HR 2.60 (2.31–2.93) for <30 ml/min/1.73 m2, both p < 0.0001). However, it was also greater amongst patients with higher eGFRs (HR 1.29 (1.19–1.41) for 90–119; HR 2.63 (2.16–3.21) for 120–150 ml/min/1.73 m2, p < 0.0001). Circulatory disease was the main cause of death in patients with low eGFRs and respiratory disease/cancer, in patients with high eGFRs. Conclusions: As a marker of mortality, both low and high eGFRs are equally predictive of increased mortality in community patients, exhibiting a ‘U’ shaped curve. Thus, current CKD guidelines which recommend inaction or even the non-reporting of eGFR values greater than 60–90 ml/min/1.73 m2 may not identify patients who are at an equally high risk of dying as those where intervention is recommended.