Elderly Greek-born Australians (GA) consistently show lower rates of all-cause and CVD mortality compared with Australian-born. Paradoxically, however, this is in spite of a higher prevalence of CVD risk factors. This paper reviews the findings from the Food Habits in Later Life (FHILL) study, other studies on Greek migrants to Australia and clinical studies investigating dietary mechanisms which may explain the "morbidity mortality paradox". The FHILL study collected data between 1988 and 1991 on diet, health and psycho-social variables on 818 people aged 70 and over from Sweden, Greece, Australia (Greeks and Anglo-Celts), Japan and were followed up for 5-7 years to determine survival status. The FHILL study was the first to develop a score which captured the key features of a traditional plant-based Mediterranean diet pattern (MDPS). A higher score improved overall survival in both Greek and non-Greek elderly reducing the risk of death by 50% after 5-7 years. Of the 5 cohorts studied, elderly GA had the lowest risk of death, even though they had the highest rates of obesity and other CVD risk factors (developed in the early years of migration with the introduction of energy dense foods). GA appeared to be "getting away" with these CVD risk factors because of their continued adherence in old age to a Mediterranean diet, especially legumes. We propose that the Mediterranean diet may, in part, be operating to reduce the risk of death and attenuate established CVD risk factors in GA by beneficially altering the gut microbiome and its metabolites.