Objective. To investigate the association between edentulism and obesity in the Swedish population aged 55-84 years over a 22-year period as a result of changes in health and socio-economic factors. Material and methods. Subjects aged 55-84 years (n = 16 416) were randomly sampled from the Swedish population by Statistics Sweden on four occasions (1980-81, 1988-89, 1996-97 and 2002). Trained interviewers collected information about dental status and anthropometric, demographic, socioeconomic, lifestyle and health-related factors. Statistical analyses were based on logistic regression models. Results. Edentulism decreased from 43% to 14% in the age group 55-84 years from 1980 to 2002, and the proportion of subjects with removable dentures decreased from 68% to 33%. In the age group 55-74 years, the proportion of subjects with low education decreased from 60% to 28%, and the proportion of obese subjects (body mass index >= 30 kg/m(2)) increased from 9% to 15%. In women aged 55-74 years, the association between obesity and edentulism, adjusted for health, lifestyle and socioeconomic factors, was significant in all surveys, and the odds ratio for obesity changed from 1.64 (95% confidence interval 1.18-2.27) in 1980 to 3.17 (95% confidence interval 1.69-6.18) in 2002. In men, the association was weaker and was significant only in the sample that combined all surveys and included individuals aged 55-84 years. Conclusion. The study indicated an association between edentulism and obesity, which was most obvious in women aged 55-74 years.

Objective: The purpose of this study was to explore the association between body composition in the elderly and subsequent changes in muscle strength during aging.Methods: This was a longitudinal study with a 5-y follow-up. Eighty-seven men (n = 38) and women (it = 49) from a random sample of 75-y-old subjects in the Goteborg part of the Nordic Research on Aging study who were investigated at ages 75 and 80 y and were free from any major diseases at baseline were included. Body composition was estimated from bioelectrical impedance. The maximal isometric strengths of handgrip, ami flexion, and knee extension were measured on the side of the dominant hand while a subject was in a sitting position in an adjustable dynamometer chair.Results: Fat-free mass decreased significantly (P < 0.001) in both sexes, but more in men. Percentage of body fat increased only in men (P < 0.05). Body height decreased in both sexes, but more in women (P < 0.001). Declines in muscle strengths were evident for all muscle groups in both sexes but more prominent in men. It was observed that body composition status at baseline, measured as fat-free mass and fat-free mass index, was a statistically significant predictor for decline in muscle strength, particularly in the extremities.Conclusion: Fat-free mass at age 75 y was associated with lower 5-y decline in muscle strength. This finding underscores the potential importance of fat-free mass for maintaining functional ability during aging. (c) 2009 Published by Elsevier Inc.

Background: Within a larger study of social network and nutrition, we investigated measurements of nutritional status and health related quality of life. Objective: To relate a well-established questionnaire of nutritional status (MNA) to a likewise well-established questionnaire of health related quality of life (SF-36) in community dwelling, free-living and, healthy 70-75 years old persons. Design: Before an interview, the MNA and SF-36 questionnaires were filled in by 128 participants from a sample of 262 subjects. Results: The MNA worked well as a measurement in this sample. Many MNA aspects correlated with the SF-36 scales. The correlations between MNA total score and the eight SF-36 scales varied from .27 to .62. Discussion: This correlation was partly due to the fact that MNA has questions of health but also to the fact that there is an empirical relation between nutrition and health. Conclusion: The MNA measurement is applicable to a healthy, free-living elderly population and parts of the MNA can be interpreted as measurements of health related quality of life. Low values of SF-36 could also be used as predictors of risk of malnutrition, although further studies are required to confirm this result.

OBJECTIVE: Secular increases in obesity have been widely reported in middle-aged adults, but less is known about such trends among the elderly. The primary purpose of this paper is to document the most recent wave of the obesity epidemic in population-based samples of 70-y-old men and women from Göteborg. Additionally, we will investigate the influences of physical activity, smoking and education on these secular trends. POPULATIONS AND METHODS: Five population-based samples of 3702 70-y-olds (1669 men and 2033 women) in Göteborg, Sweden, born between 1901 and 1930, were examined in the Gerontological and Geriatric Population Studies (H70) between 1971 and 2000. Cohort differences in anthropometric measures were the main outcomes studied. Physical activity, smoking habits and education were assessed by comparable methods in all cohorts. Subsamples of the women in the latest two cohorts (birth years 1922 and 1930) were also part of the Prospective Population Study of Women in Göteborg. In these women, it was possible to examine body mass index (BMI) and waist-to-hip circumference ratio (WHR) longitudinally since 1968. RESULTS AND CONCLUSIONS: Significant upward trends were found for height, weight, BMI, waist circumference (WC), WHR, prevalence of overweight (BMI≥25 kg/m 2 ) and obesity (BMI≥30 kg/m 2 ) across cohorts in both sexes. In 2000, 20% of the 70-y-old men born in 1930 were obese, and the largest increment (almost doubling) had occurred between the early 1980s and the early 1990s. In 70-y-old women the prevalence of obesity was 24% in 2000, a 50% increase compared to the cohort born 8 y earlier. BMI increased over time in all physical activity, smoking and education groups, with the exception of never-smoking men. Although 70-y-old women in 2000 were heavier than cohorts examined 8 y previously, data from the women studied longitudinally revealed that these differences were already present in earlier adulthood. In conclusion, the elderly population is very much part of the obesity epidemic, although secular trends in BMI were detected slightly earlier in men than in women. The health implications of these secular trends should be focused on in future gerontological research.

OBJECTIVETo develop a prediction equation for percent body fat (BF%) from skin fold thickness in a random sample of free-living 75-year-olds and to compare with published equations for the elderly.SETTINGGoteborg part of the Nordic Research on Aging (NORA) study in 1991/92.SUBJECTS & METHODSAnthropometric measurements were performed in 286 participants (125 males,161 females) and body composition was estimated by four compartment model (4C) from total body water (TBW) and total body potassium (TBK) in every alternative subjects. BF % predicted from skin fold thickness (BFSF) was validated against BF % estimated from 4C model (BF4C) and was compared with BFSF and BF % predicted from equations by Durnin and Womersley, 1974 (BFDW), Deurenberg et al.,1989 (BFPD) and, Visseretal.,1994(BFMV), respectively.RESULTSThe BFSF correlated well with BF4C (r = 0.86, SEE = 4.05). Compared to BF4C, both BFPD and BFMV over estimated BF % by 3.26 and 3.63 in males and 9.56 and 9.23 in females, respectively. BFDW underestimated BF % in females by 1.06 and overestimated by 3.12 in males. BFSF showed best agreement with BF4C, where the 95 % of differences lie between +/- 7.4 in males and +/- 8.7 in females, respectively.CONCLUSIONBF % predicted from skin fold thickness correlated well with estimated BF % in 75-year-olds. Different prediction equations gave different values for BF % and population-specific prediction equations seem preferable in the elderly population.

OBJECTIVE:To investigate the role of obesity in general and waist circumference (WC) and BMI in particular as risk factors for 15-year incidence of coronary heart disease (CHD) in the elderly. RESEARCH METHODS AND PROCEDURES:This prospective study was based on 1597 (737 males and 860 females) 70-year-olds free from CHD and participants of three birth cohorts examined in 1971 to 1972 (Cohort I), 1976 to 1977 (Cohort II), and 1981 to 1982 (Cohort III) at Göteborg, Sweden. Fifteen-year incidence of CHD (fatal and nonfatal) was ascertained from follow-up examinations and registers. Relative risk (RR) for first ever CHD in reference to the lowest quartiles of WC and BMI was calculated from Cox regression. RESULTS:In males, RRs for CHD in the highest WC and BMI quartiles were 1.36 [95% confidence interval (CI) 1.00 to 1.85] and 1.42 (95% CI 1.04 to 1.92), respectively, after adjustment for cohorts, smoking habits, diabetes, systolic blood pressure, and total cholesterol. In men, the risk associated with WC was independent of BMI. Neither WC nor BMI was related to CHD risk in females. After exclusion of first 5-year all-cause deaths, the adjusted RRs in the highest WC and BMI quartiles in males were 1.47 (95% CI 1.06 to 2.04) and 1.42 (1.04 to 1.92), respectively. In females, a significantly higher RR of 1.41 (95% CI 1.02 to 1.94) was observed in the second BMI quartile only after such exclusions. DISCUSSION:WC, an indicator of both central and general obesity, appears to be a stronger predictor of CHD than BMI in elderly males, but in females, obesity was not a risk factor for CHD.

The H70 longitudinal study of aging, Göteborg, Sweden is used to empirically test the compression of morbidity theory advanced by. We reconceptualize compression as postponement of morbidity in the sense of decreasing amounts of illness for increasingly long life spans. Operationally, morbidity is defined as the average number of hospital days in the last year of life. The date of death and the date of 1-year prior to death define the risk period. The linear regression model with age at death, age at death squared, year of birth, and sex are statistically significant with the oldest having the fewest hospital days. The findings offer partial support for the compression of morbidity theory.

OBJECTIVE: We evaluated the performance of different prediction equations to estimate fat-free mass (FFM) from bioelectrical impedance analysis (BIA) in the elderly.METHODS: This study was based on 106 (51 male and 55 female) free-living 75-y-old subjects who participated in the Goteborg part of the Nordic Research on Ageing (NORA) study during 1991 and 1992. FFM predicted from BIA (FFMGOT) was validated against FFM estimated from measurements of total body water and total body potassium (FFMREF). FFM was calculated from BIA prediction equations for the elderly developed by Deurenberg et al. (FFMWAG) and Roubenoff et al. (FFMFHS). FFM also was calculated from an equation developed in subjects with a wide age range by Kyle et al. (FFMGEN). Bland-Altman analysis was performed to compare FFMREF with FFMGOT,, FFMWAG, FFMFHS, and FFMWAG respectively. FFMGOT also was compared with FFM derived from these published equations.RESULTS: Compared with FFMREF, the FFMFHS and FFMWAG underestimated FFM by 2.6 and 7.9 kg in males and 4.2 and 9 kg in females, respectively. The FFMGEN underestimated FFM in females by 1.3 kg but not in males (mean difference, -0.04 kg). FFM calculated from the BIA equation developed in this population (FFMGOT) neither underestimated nor overestimated FFM as compared with FFMREF, as expected. The differences between FFMGOT and FFMs predicted from these equations were of the same magnitude as that observed with FFMREF.CONCLUSION: Different prediction equations produced different values for FFM. The age-specific equations developed in other populations underestimated FFM, whereas FFMGEN produced an unbiased estimate of FFM in males but not in females. Thus, the BIA prediction equation needs to be developed and validated in the population under study. (C)Elsevier Inc. 2003.

Objective: To develop a bioelectrical impedance (BIA) prediction equation for fat-free mass (FFMBIA) and present reference values of FFM and body fat (BF) for healthy Swedish elderly from population-based representative samples. Subjects: This study is based on 823 (344 males, 479 females) participants from two systematic samples of birth cohorts in Göteborg aged 70 (cohort H70V, 201 males and 299 females) and 75 (cohort NORA75, 143 males and 180 females). Methods: Body composition was measured with BIA (BIA-101, RJL system, Detroit) in both cohorts and was estimated by a four-compartment (4C) model from total body water (TBW) and total body potassium (TBK) in a sub-sample of the NORA75 cohort. The FFMBIA was validated against the FFM from the 4C model (FFM4C). Results: The FFMBIA correlated well with FFM4C (r=0.95, SEE=2.64 kg). The FFMBIA (kg) in 70-y-old males and females were 58.5±5.4 and 43.4±4.4, and for 75-y-old males and females were 56.1±4.7 and 42.5±4, respectively. The body fat in kg (FM) among 70-y-old males and females were 25.2±8.1 and 25.7±8.4, and for 75-y-old males and females were 21.7±7.1 and 22.8+7.2, respectively. The percent body fat (BF%) among 70-y-old males and females were 29.5±5.8 and 36.3±6.4, and for 75-y-old males and females were 27.3±6 and 34.1±6.1, respectively. Conclusion: The FFM, FM and BF% from this study might be used as reference values for Swedish elderly aged 70 and 75 y. Sponsorship: See acknowledgement.

OBJECTIVES: To investigate waist circumference (WC) and body mass index (BMI) at age 70 as risk factors for stroke.

Objective: To examine the relationship between body mass index (BMI) at age 70, weight change between age 70 and 75, and 15 y mortality. Design: Cohort study of 70-y-olds. Setting: Geriatric Medicine Department, Göteborg University, Sweden. Subjects: A total of 2628 (1225 males and 1403 females) 70-y-olds examined in 1971–1981 in Gothenburg, Sweden. Results: The relative risks (RRs) for 15 y mortality were highest in the lowest BMI quintiles of males 1.20 (95% CI 0.96–1.51) and females 1.49 (95% CI 1.14–1.96). In non-smoking males, no significant differences were observed across the quintiles for 5, 10 and 15 y mortality. In non-smoking females, the highest RR (1.58, 95% CI 1.15–2.16) for 15 y mortality was in the lowest quintile. After exclusion of first 5 y death, no excess risks were found in males for following 5 and 10 y mortality across the quintiles. In females, a U-shaped relation was observed after such exclusions. BMI ranges with lowest 15 y mortality were 27–29 and 25–27 kg/m2 in non-smoking males and females, respectively. A weight loss of≥10% between age 70 and 75 meant a significantly higher risk for subsequent 5 and 10 y mortality in both sexes relative to individuals with ‘stable’ weights. Conclusion: Low BMI and weight loss are risk factors for mortality in the elderly and smoking habits did not significantly modify that relationship. The BMI ranges with lowest risks for 15 y mortality are relatively higher in elderly. Exclusion of early deaths from the analysis modified the weight–mortality relationship in elderly males but not in females. Sponsorship: See Acknowledgements. European Journal of Clinical Nutrition (2001) 55, 482–492

BACKGROUND:Body size in elderly adults is partly due to aging and partly to secular trends. This study describes secular trends in three anthropometric measures (i.e., height, body weight [BW], and body mass index [BMI]) of 70-year-olds over a period of 21 years and their relation to social and lifestyle factors.METHODS:A total of 3128 70-year-olds from four birth cohorts born between 1901 and 1922 in Gothenburg, Sweden, were examined between 1971 and 1992 in the Geriatric Medicine Department, Göteborg University. Trends in anthropometric measures were examined by permutation test. Influence of the subjects' birth year, physical activity, smoking habits, and education on anthropometric measures were investigated by multiple linear regression.RESULTS:Individuals in later-born cohorts were found to be 1 to 2 cm taller and 1.5 to 6.3 kg heavier than earlier-born cohorts. For BMI, a positive trend was significant only in 70-year-old male participants. "Year of birth" was a positive predictor for BW (p <.001) and BMI (p <.001) in male participants and for height (p <.05) and BW (p <.01) in female participants. Physical inactivity was a positive (p <.01) and "current smoking" a negative (p <.001) predictor for BMI in both sexes. "More than basic education" was a positive predictor for height (p <.001) in both sexes and a negative predictor for body weight (p <.01) and BMI (p <.001) in female participants only.CONCLUSIONS:Trends of increasing height, BW, and BMI were found among the Swedish elderly participants. This may be partly due to differences in smoking habits, physical activity, education, food habits, childhood nutrition, and living conditions between the cohorts.

Objective : To describe longitudinal changes in height and body weight between the ages of 70 and 95 y. Design : Longitudinal cohort study with representative sample of 70-y-olds. Setting : Department of Geriatric Medicine, Göteborg University, Sweden. Subjects : 449 males and 524 females, aged 70 y, living in Göteborg were examined in 1971–72 and this study population participated on 11 occasions during a 25-year follow-up. Results : Mean height decreased 4 and 4.9 cm in males and females respectively and the trend was significant between the ages of 70 and 95 y in both sexes. Between 70 and 75 y of age, a significant difference was found between quintiles of body height where in the highest quintile height was lowered by 0.4 and 0.3 cm/y, in males and females respectively, and in the lowest quintile by 0.1 cm/y in both sexes. Mean body weight decreased 3.2 and 5.1 kg in males and females respectively, from age 70 to 95 y. The trend was significant over 22 and 20 y for males and females, respectively. Between the ages of 70 and 80 y, individuals in highest quintile of body weight decreased at a rate of 0.8 and 0.6 kg/y, three times higher than those in lowest quintile. Due to the decrease in both height and weight over time, body mass index (BMI) was less affected. Conclusion : Height, body weight and BMI decreased significantly in both sexes after age 70 y, and there was a gender difference in the trend. The results can be used as reference data for Swedish elderly and might be of importance to the understanding of anthropometry with the ageing process. Sponsorship : See acknowledgements.