Although associations between work stressors and stress-related biomarkers have been reported in cross-sectional studies, the use of single time measurements of work stressors could be one of the reasons for inconsistent associations. This study examines whether repeated reports of work stress towards the end of the working career predicts allostatic load, a measure of chronic stress related physiological processes. Data from waves 2 to 6 of the English Longitudinal Study of Ageing (ELSA) were analysed, with a main analytical sample of 2663 older adults (aged 50+) who had at least one measurement of effort-reward imbalance between waves 2-6 and a measurement of allostatic load at wave 6. Cumulative work stress over waves 2-6 were measured by the effort-reward imbalance model. ELSA respondents who had reported two or more occasions of imbalance had a higher (0.3) estimate of the allostatic load index than those who did not report any imbalance, controlling for a range of health and socio-demographic factors, as well as allostatic load at baseline. More recent reports of imbalance were significantly associated with a higher allostatic load index, whereas reports of imbalance from earlier waves of ELSA were not. The accumulation of work related stressors could have adverse effects on chronic stress biological processes.