Dietary patterns and health in the elderly. A north-south comparison in Europe.

The aim of this thesis was to gain more insight into the role of dietary intake as a determinant of different aspects of health, e.g. cognitive function, functional status, self-rated health and mortality. Dietary patterns, obtained from cluster analysis or a healthy diet indicator (HDI), were used instead of single nutrients to investigate these associations. The results are based on different datasets. The largest part was performed in five population-based samples from three European countries participating in the Seven Countries Study: East and West Finland, Zutphen in The Netherlands, and Crevalcore and Montegiorgio in Italy. Around 1970, this dataset consisted of more than 3000 men aged 50-69 years. Around 1990 more than 1000 of the men aged 70-89 years participated in the follow-up. Another dataset used, comprised the information of 272 men and women aged 64-87 years in 1971, from a general practice in Rotterdam, The Netherlands.The first part of the thesis concerned the use and the comparison of cluster analysis and the healthy diet indicator. Using cluster analysis, groups with different dietary patterns can be discerned, which have different profiles of risk factors for cardiovascular diseases. Furthermore, an association was observed with 17-year all-cause mortality among men. The HDI, a score which is based on uniform cut-off values derived from recommendations of a WHO study group, appeared to be less dependent from the dataset under study. It was also associated with 17-year all-cause mortality in men but not in women.The remaining part of the thesis concerned the associations of dietary intake with several determinants of health status at old age in the combined datasets of Finland, Italy, and The Netherlands. Cross- sectional analysis of the dietary intake at old age (collected around 1990) indicated that the general patterns of dietary intake of the different cultures could still be recognised. A retrospective comparison with the dietary intake at middle age (around 1970) lead to the conclusion that since then, the dietary pattern of the Finnish and Dutch cohorts had changed slightly into the direction of a more healthy diet, e.g. less atherogenic, while the diet of the Italian men had remained Mediterranean. A healthy diet, as calculated by the HDI, was associated with a reduced 20-year allcause and cardiovascular mortality in the three different cultures. Furthermore, there was a tendency towards a better cognitive function among the elderly men with a better HDI in four out of five cohorts. Physical disabilities were accompanied by lower energy intake in elderly men in all three countries. This association was largely, but not completely, due to a lower physical activity. Self-rated health seemed to be a very culture dependent variable, since more than 80% of the elderly men in Finland felt moderately or not healthy, while this was only 20% and 10% in Italy and The Netherlands respectively. Only in Italy, a low self-rated health was associated with a lower energy intake. Since the studies on cognitive funtion, physical disabilities and self-rated health were designed crosssectional it cannot be decided whether the observed associations were causal or not.In conclusion, the studies presented in this thesis underline the importance of dietary intake in the maintenance of health at old age (e.g. mortality, cognitive function, functional status, self-rated health). Instead of calculating the associations of several health outcomes with single nutrients, a dietary pattern approach was used. The results of this international study on diet and health suggests that a healthy dietary pattern might have a beneficial effect on health status, but longitudinal studies are needed to confirm this assumption