Objective: The purpose of this study was to compare dietary practices among different birth cohorts of 70-year-old Swedes, who were examined between 1971 and 2000. Setting: Göteborg, Sweden. Design: Four population-based samples of 1360 70-year-olds, born in 1901, 1911, 1922 and 1930, have undergone health examinations and dietary assessments over a period of almost three decades. One-hour diet history (DH) interviews were conducted in 1971, 1981, 1992 and 2000 with a total of 758 women and 602 women. The formats and contents of the dietary examinations were similar over the years. Statistical analysis of linear trends was conducted, using year of examination as the independent variable, to detect secular trends in food and nutrient intakes across cohorts. Results: At the 2000 examination, the majority of 70-year-olds consumed nutritionally adequate diets. Later-born cohorts consumed more yoghurt, breakfast cereals, fruit, vegetables, chicken, rice and pasta than earlier-born cohorts. Consumption of low-fat spread and milk also increased, along with that of wine, light beer and candy. In contrast, potatoes, cakes and sugar were consumed less in 2000 than in 1971. The ratio of reported energy intake to estimated basal metabolic rate did not show any systematic trend over time in women, but showed a significant upward trend in men. Conclusions: The diet history method has captured changes in food selections in the elderly without changing in general format over three decades. Dietary quality has improved in a number of ways, and these findings in the elderly are consistent with national food consumption trends in the general population.
KeywordsElderly Diet Cohort Gerontology Secular trends Life expectancy among the elderly has changed over recent decades from 72.3 years (men) and 78.1 years (women) in 1971 to 77.1 and 81.9 years, respectively, in 2000 1 . Thus, the elderly constitute a rapidly growing segment of the population in Sweden, as in most Western countries. This changing demographic structure makes it increasingly important to document dietary quality in older individuals, many of whom are at risk of nutritional deficiencies 2 . Relatively little is known about energy and nutrient intakes among free-living elderly populations. However, many hospitalised elderly patients in Sweden are undernourished, and the large variation in energy and nutrient intakes in free-living elderly populations confirms the presence of subgroups that are nutritionally at risk 2 . The main aim of the present study was to compare the diets of four population-based samples of 70-year-old Swedes who were examined between 1971 and 2000, with particular attention to the latest examination in the cohort born in 1930.
Methods
Populations sampledThe data for this study were obtained from two sources: the Gerontological and Geriatric Population Studies in Gothenburg ('H70') 3 -5 and the Population Study of Women in Gothenburg 6 -9 . Both studies have been reviewed and approved by the ethics committee of Göteborg University. Go...