Background. The prognostic value of reduced glomerular filtration rate (GFR) was examined in a community-based cohort of British women. Methods. Serum creatinine measurements were available for 90% (n = 3851) of a representative random sample of 4286 women aged 60-79 years. GFR was estimated using the Modification of Diet in Renal Disease equation. Hazard ratios (HR) were calculated using Cox regression with outcomes of all-cause and cardiovascular disease (CVD) mortality. Results. Eight hundred and thirty-two women (21.6%) had a GFR <60 ml/min/1.73 m(2). Over a median follow-up of 5.6 years, there were 318 deaths (100 CVD deaths). Women with GFR <60 ml/min/1.73 m(2) compared to all others showed only a borderline increased risk of all-cause mortality [HR 1.35 (95% confidence intervals: 0.99, 1.85)] and CVD mortality [1.34 (0.97, 1.85)]. Adjustment for conventional CVD risk factors had little impact. The association with CVD mortality was attenuated in women with pre-existing CVD [adjusted HR: 0.51 (0.24, 1.04)]. Only the subset of women without CVD at baseline were at risk for later CVD death [adjusted HR: 1.80 (1.13, 2.88)]. Conclusions. A substantial proportion of older British women have GFR <60 ml/min/1.73 m(2) without strong evidence for statistical association with all-cause mortality. The effect on CVD mortality is partly explained by existing CVD and its risk factors. GFR measurement appears only to play a useful role in the subset of older women without pre-existing CVD who are at higher risk of premature CVD death.