Background: Measures of biological age have not been proven to predict mortality. This study examines whether measuring biological age improves the prediction of mortality. Methods: Prospective study from 1981 to 2001 of 397 male London Civil Servants. Two indices of biological ageing were calculated. Results: 60 men died and both indices of biological ageing were related to survival. In a model that mutually adjusted for both chronological and biological age, biological age using index one was statistically significant with a hazard ratio (HR) of 1.11 per year of age (95% confidence interval 1.01 - 1.21, P=0.03). The useful components of the measures of biological ageing were systolic blood pressure (HR 1.31 for 1SD), albumin, and, to a lesser degree, Erythrocyte Sedimentation Rate (ESR). Greying of the hair, skin inelasticity, arcus senilis, and baldness were not predictors of mortality as measured by our methods. Similarly serum cholesterol, creatinine, calcium and urate could be excluded. A modified index was developed including systolic pressure, ESR, urea, albumin, and bilirubin and had a sensitivity of 78% and specificity of 51% in predicting subjects who died. Conclusion: This study represents ‘proof of principle’ in demonstrating the utility and validity of measuring biological age. The modified index needs to be tested prospectively.