Dietary patterns by reduced rank regression predicting changes in obesity indices in a cohort study: Tehran Lipid and Glucose Study.

OBJECTIVE: To examine the association between dietary patterns and obesity indices (BMI, WC, WHR) among Tehranian adults in a 6-year follow-up study. METHODS: Within frame of a cohort study in Tehran (mean follow up 6.6+/-0.9 years), 141 adults were recruited with: two 24 hour dietary recalls at the beginning, as well as obesity indices at the beginning and end of the study period. Dietary intakes were converted into grams of intakes of food items and categorized into 16 groups. Reduced rank regression analysis derived five patterns with total and polyunsaturated-to-saturated fat intake, cholesterol, fiber and calcium intake as response variables. Factors (dietary patterns) were generated retaining a corresponding factor loading > or = |0.17| on the food groups. Changes in obesity indices were scrutinized within quintiles of factor scores. RESULTS: There were high loadings on refined carbohydrates, whole grain, starchy vegetables, other vegetables, red and refined meat, saturated/trans fat, and egg for the first factor named "traditional". All obesity indices had increasing trend across quintiles of pattern score. The fifth pattern (namely egg pattern) had high loading for eggs, salty snacks, as well as fruits and dry fruits, and negative loadings for red and processed meat, saturated and trans fat, plant oils, and dairy products. This pattern showed increasing trends for WC and WHR after adjustment for potential confounders. Other patterns showed non-significant trends for obesity indices. CONCLUSIONS: The results were indicative of a traditional pattern which is dominated in the Tehran region and associated with increase in obesity indices.

Reproducibility and relative validity of food group intake in a food frequency questionnaire developed for the Tehran Lipid and Glucose Study.

OBJECTIVE: To examine the validity and reproducibility of food groups in the semi-quantitative food frequency questionnaire (FFQ) developed for the Tehran Lipid and Glucose Study (TLGS). METHODS: To evaluate the reproducibility of food groups included in the FFQ, 132 subjects (61 men and 71 women) aged 20 years or older twice completed a 168-item FFQ (FFQ1, FFQ2), with a 14-month interval between FFQ1 and FFQ2. Over the 1-year interval, 12 dietary recalls (DRs) were collected (1 each month) to assess the validity of the FFQ. Seventeen food groups were derived from the FFQ based on methods described in previous studies. Age-adjusted and deattenuated Spearman correlation coefficients were used to assess the validity of the FFQ. RESULTS: The mean (SD) age and body mass index of subjects were 35.5 (16.8) years and 25.5 (5.2) kg/m(2), respectively. Validity correlation coefficients ranged from 0.03 (liquid oil) to 0.77 (simple sugars) in men (median, 0.44), and from 0.12 (snacks) to 0.79 (simple sugars) in women (median, 0.37). The energy-adjusted intraclass correlation coefficient, which reflects the reproducibility of the FFQ, was 0.51 in men and was highest for tea and coffee (0.91); in women it was 0.59 and was highest for simple sugars (0.74). The highest percentage of complete agreement and disagreement was observed for snacks and desserts (60.6%) and potatoes and dairy products (12.8%), respectively, in men, and tea and coffee (62.9%) and legumes (15.7%) in women. CONCLUSIONS: The FFQ that was designed for the TLGS was found to be reliable and valid for assessing the intake of several food groups.

Reliability and relative validity of an FFQ for nutrients in the Tehran lipid and glucose study.

OBJECTIVE: To describe the relative validity and reliability of the FFQ used for assessing nutrient intakes of participants in the Tehran Lipid and Glucose Study (TLGS). DESIGN: A total of 132 subjects (sixty-one males and seventy-one females) were included in the study. Dietary data were collected monthly by means of twelve 24 h dietary recalls (24hDR). Subjects completed two, 168-item semi-quantitative FFQ. Blood and urine samples were taken every season for measurement of plasma biomarkers and urinary N and K. RESULTS: Mean age and BMI of the participants were 35.5 (sd 16.8) years and 25.5 (sd 5.2) kg/m2, respectively. The mean energy-adjusted and deattenuated correlation coefficients for overall nutrient intake between the 24hDR and FFQ2 were 0.44 and 0.37 in 35-year-olds, respectively, and for individual nutrients ranged from 0.24 to 0.71 in men (mean r = 0.53) and from 0.11 to 0.60 in women (mean r = 0.39). The mean energy-adjusted reliability coefficients varied from 0.48 in 35-year-olds, and ranged from 0.41 to 0.79 in men (mean r = 0.59) and from 0.39 to 0.74 in women (mean r = 0.60). The FFQ2 and 24hDR produced exact agreement rates ranging between 39.6 % and 68.3 % in men and between 39.6 % and 54.1 % in women. The ranges of questionnaire validity coefficients, with the sample correlation between the questionnaires and biochemical marker as the lower limit and the estimate obtained by the method of triads as the upper limit, were 0.21-0.56 (protein) and 0.37-0.61 (K). CONCLUSIONS: The FFQ developed for the TLGS has reasonable relative validity and reliability for nutrient intakes in Tehranian adults.