Sensitive and specific biomarkers of ageing are needed to evaluate interventions to extend health span. However, there is growing evidence that information provided by candidate biomarkers may change with age itself. Little is yet known about the value of candidate biomarkers in those over 85 years, currently the fastest growing population sub-group in many countries. This study assessed a large panel of candidate biomarkers in a cohort of 85 years old by studying comparative associations with health status. Using a cross-sectional sample of 852 individuals aged 85, we performed uni- and multi-variable analyses of associations between 74 candidate biomarkers and 4 health-status measures: viz. multi-morbidity, cognitive impairment, disability and proximity to death as measured by mortality within 1.5 years. We defined as most informative any measures that were significantly associated with at least two of the health-status measures in multivariable analyses in this age group. 10 out of 74 tested candidates fulfilled this criterion, while several proposed biomarkers of ageing, notably inflammation and immune risk markers and telomere length, did not. As future data accrues on health outcomes within the cohort, it will become possible also to evaluate the predictive value of these and others of the candidate biomarkers.