Age at disease onset, a familiar concept in human medicine, has not been commonly discussed in relation to diseases of wildlife. Furthermore, while age has been anecdotally linked with disease susceptibility in aquatic animals (e.g. in aquaculture), little attention has been devoted to health of specific cohorts in wild aquatic animal populations. In an attempt to refine current approaches to marine environmental monitoring programmes that utilise fish diseases as a sentinel for ecological status, we have investigated age at onset for 10 disease variables in the flatfish Limanda limanda, a commonly utilised species in such programmes in Europe. By comparing the prevalence of these variables (including several associated with liver neoplasia) in different age classes from the same sites, we have demonstrated an increased propensity for 'harm' (a cumulative measure of multiple disease variables) with age. Furthermore, a comparison of age-matched cohorts from geographically distinct offshore marine sites revealed that this tendency for increased harm differs in magnitude between sites. In these scenarios, the age at onset for key disease variables (such as pre-neoplastic and neoplastic pathologies of the liver) also differs between sites, with some locations containing populations with apparently increased susceptibility in younger age classes. We conclude that while age is an important variable when assessing fish population health status, it does not explain all of the differences in disease prevalence observed in natural populations, and further, that age-matched cohorts from certain populations have a higher propensity for disease than fish of the same age from other sites. The basis for these differences in susceptibility is discussed in relation to the prevailing burden of anthropogenic contaminants and to other natural factors (such as population genetics and migration) inherent in sampled populations.