| Objective:1. To evaluate the reproducibility and validity of a simple food-frequency questionnaire used in dietary patterns studies.2. To identify dietary patterns and the influencing factors.3. To investigate the single and joint associations of dietary patterns and physical activities with total body fat proportions measured by dual-energy X-ray absorptiometry among middle-aged and elderly people.4. To investigate the single and joint associations of dietary patterns and physical activities with the risk of metabolic syndrome (MS) among middle-aged and elderly people.Subject and Method:1. The simple food-frequency questionnaire (FFQ25) was administered twice in94Shanghai community residents aged45to85years old for reproducibility assessment by comparing food intakes and the correlation coefficient. The comparing and correlation analysis of food intakes, energy and nutrient intakes were conducted. The dietary patterns analysis of FFQ25, whole food-frequency questionnaire (FFQ146) and3d24h dietary records (DR) were administered in these people for validity assessment.2. A total of2325individuals45to85years of age participated in a population-based cross-sectional survey in Shanghai. The information on participants’socio-demographic status, food intakes and physical activities were collected via questionnaires. Anthropometric measurements including height, weight, waist circumference, hip circumference, blood pressure, were measured. FBG, PBG, HbAlc, triglyceride, total cholesterol, high-density lipoprotein, low-density lipoprotein and uric acid were measured in these people. Total body fat proportions were measured by dual-energy X-ray absorptiometry. Dietary patterns were derived from FFQ25using factor analysis. The differences of socio- demographic status, energy and macro nutrient intakes, anthropometric measurements and blood biochemical variables were compared between dietary patterns. Multivariate logistic stepwise regression and multivariate linear stepwise regression were used to find the influencing factors of the dietary patterns and factor scores. Correlation analysis and partial correlation analysis were used to analyze the associations between anthropometric measurements, blood biochemical variables and the total body fat proportions. Multivariate linear regression was used to estimate the association between dietary patterns, physical activities and the total body fat proportions. The agreement of the three diagnostic criteria for MS was tested by Kappa test. Multivariate logistic regression was used to estimate the association between dietary pattern, physical activities and the risk of MS.Results:1. The difference of food intakes between the two FFQ25was no significant (P>0.05) and the correlation coefficients between the two FFQ25ranged from0.56to0.87(P<0.01). The correlation coefficients of food intakes between FFQ25, FFQ146, and DR ranged from0.31to0.72(P<0.01). The correlation coefficients of energy and nutrient intakes between FFQ25, FFQ146, and DR ranged from0.34to0.81(P<0.05) and the correlation coefficients of factor loadings ranged from0.54to0.77(P<0.01). In these three dietary patterns, there were good agreements between the factor scores from FFQ25and FFQ146, FFQ25and DR by Bland-Altman analysis.2. Three evident dietary patterns were derived by factor analysis in these2325participants:Healthy (40.4%of total population), Traditional (34.4%) and Western (25.2%).The Healthy pattern was characterized by high consumptions of dairy products, dark-color vegetables, poultry, eggs, soya products, river fishes, rice, white-color vegetables and fruits. The Traditional pattern was characterized by high consumptions of soya products, rice, red meat, dark-color vegetables, gruel, flour, white-color vegetables and stuffing. The Western pattern was characterized by high consumptions of red meat, poultry, dairy products, eggs, fruits, sea products, desserts, dark-color vegetables and nuts.3. After adjusting for age, gender and total energy, the intakes of fat and the ratio of energy from fat in Western pattern was significantly higher than those in Healthy pattern (P<0.05) and the intakes of carbohydrates and the ratio of energy from carbohydrates in Western pattern is significantly lower than those in Healthy pattern (P<0.05). The intakes of protein and carbohydrates, ratio of energy from protein and carbohydrates were positively associated with the Healthy factor scores (P for trend<0.01). The intakes of fat and ratio of energy from fat were adversely associated with the Healthy factor scores (P for trend<0.01).4. Comparing with the Healthy pattern, BMI, waist circumference, hip circumference and uric acid were significantly higher in the Western pattern (P<0.05). Comparing with the Healthy pattern, BMI, waist circumference, hip circumference, WHR, SBP, DBP, FBG, triglyceride, and uric acid were significantly higher in the Traditional pattern (P<0.05)5. The differences of vigorous-intensity activity, low-intensity activity, walking and sitting between dietary patterns were significant (P<0.05), while the differences of total physical activity and moderate-intensity activity between dietary patterns were insignificant.6. The main influencing factors of Healthy factor scores were gender, smoking status, drinking status and total physical activity. The main influencing factors of Western factor scores were age, monthly household income, smoking status and drinking status. The main influencing factors of Traditional factor scores were age, monthly household income, education, smoking status and total physical activity.7. Total body fat proportions were associated with many anthropometric measurements and blood biochemical variables. After adjusting for age, gender and BMI, the partial correlation coefficients were still significant (P<0.05), but all of them were very small.8. Dietary patterns were independent influencing factors of total body fat proportions. Comparing with the Healthy pattern, Western pattern and Traditional pattern had higher total body fat proportions(p=1.30, P<0.001and P=0.74, P=0.004). Total body fat proportions were adversely associated with the Healthy factor scores (P for trend=0.041), while it were positively associated with the Western factor scores (P for trend=0.014)9. Physical activities were independent influencing factors of total body fat proportions. Total body fat proportions were adversely associated with the total physical activity and walking (P for trend P<0.05), while it were positively associated with sitting (P for trend<0.001). The interaction for total body fat proportions was found between dietary patterns and sitting (P=0.009), but the interaction for total body fat proportions was not found between dietary patterns with other physical activity variables.10. The MS prevalence rate was24.7%,27.1%and23.8%in CDS2004, IDF2005and NCEP-ATPIIII respectively. The diagnoses agreement of Kappa was0.58,0.61and0.84respectively. All these kappa value were significant (P<0.001).11. BMI, waist circumference, hip circumference, blood pressure, FBG, PBG, HbAlc, triglyceride, total cholesterol and uric acid in MS group were higher than that of Non-MS group (P<0.01), while HDL in MS group were lower than that of Non-MS group (P<0.01). The difference of LDL between two groups was insignificant.12. Dietary patterns were independent influencing factors of the risk of MS. Comparing with the Healthy pattern, Western pattern was associated with greater odds of having MS (odds ratio1.44,95%confidence interval1.12-1.85), while there was no association between Traditional pattern and the risk of MS (odds ratio1.11,95%confidence interval0.87-1.40). The western pattern was associated with greater odds of having overweight or obesity, hypertriglyceridemia, elevated blood pressure and dyslipidemia (odds ratio>1), but were insignificant. The risk of MS was positively associated with the Western factor scores (P for trend P=0.002), while it were not associated with Healthy factor scores and Traditional factor scores.13. Physical activities were independent influencing factors of the risk of MS. Total physical activity, low-intensity activity, walking were associated with lower odds of having MS (P<0.05), while sitting was associated with greater odds of having MS (P=0.008). The interactions for the risk of MS were not found between dietary patterns and physical activities.Conclusion:1. FFQ25showed reasonable reproducibility and validity of the major dietary patterns. It was very easy and convenient. It could be applied in dietary patterns studies in elderly people in cities.2. Three dietary patterns were derived among middle-aged and elderly people in Shanghai:Healthy, Traditional and Western. Healthy pattern accounted for the largest proportion and Western pattern accounted for the smallest proportion. There were significantly differences between dietary patterns in energy and macro nutrient intakes, anthropometric measurements, blood biochemical variables and physical activities.3. Healthy pattern was inversely associated with total body fat proportion, while Western pattern and Traditional pattern were positively associated with total body proportions.4. Healthy pattern was associated with lower risk of MS and Western pattern was associated with the increased risk of MS. The traditional pattern showed no association with the risk of MS. Physical activities were associated with lower risk of MS and sitting was associated with the increased risk of MS.5. Dietary patterns and physical activities were significantly associated with the risk of increased total fat proportions and MS. There were some influences for each other between dietary patterns and physical activities. Adherence to a Healthy dietary pattern and having enough physical activities were very important for the reduced risk of obesity and MS. |
PDF Full Text Request