23 Jan

Fruit, Vegetable and Fat Intake: METHODS

Data Source and Study Population

We analyzed data from the baseline assessment of Project DIRECT (Diabetes Interventions Reaching and Educating Communities Together); methods are described in detail elsewhere. Briefly, a baseline assessment was conducted in 1997 using a multistage, population-based probability sample from U.S. census files in predominately African-American neighborhoods in Raleigh and Greensboro, NC. Trained field interviewers visited each household and selected eligible persons according to a specified protocol. Individuals who were eligible were invited to complete a personal interview; the overall response rate was 87%. The total study sample included 2,310 people (2,210 African Americans, 65 whites and 35 members of other races). In this cross-sectional analysis, we evaluated only those individuals who self-identified race as African-American. Individuals who had missing data on sociodemographic (not including income) or dietary variables were excluded, yielding a final study population of 2,172 African Americans.

Study Variables and Measures

Sociodemographic variables (age, sex, education, income, marital status, employment status), health variables (health status, height, doctor recommendation of weight loss status) and behavior variables (attempting, physical activity level [active was determined by report of any physical activity in the past month]) were determined by self-report. Weight was measured, and body mass index (BMI) was calculated in kg/m2 and categorized according to National Institutes of Health guidelines: optimal/underweight <25, overweight 25-<30, obese 30-<35, extremely obese (obesity II) >35.

Fruit, vegetable and fat intake were the main focus of this analysis and were assessed using a modified version of the Block questionnaire. The original Block questionnaire (100 items) was developed into shorter screeners (22 items) to assess fat (15 items) and fruit and vegetable intake (seven items), and validated against the original version. Project DIRECT included six questions to estimate fruit and vegetable intake and 13 questions regarding fat intake that were modified from the Block short-screeners that were used in the Behavioral Risk Factor Surveillance System (BRFSS) telephone survey. Using these data, variables were calculated to estimate several dietary domains according to methods outlined for the short Block screener questionnaire: fruit and vegetable intake in daily servings and daily total and saturated fat intake in grams.

Statistical Analysis

To account for the complex survey design, all analyses were conducted using STATA Survey commands (STATA Statistical Software, Release 7.0, College Station, TX 2003). Descriptive analyses (means and frequencies) were conducted for sociodemographic, dietary, health and behavior variables.

Daily servings of fruits and vegetables were calculated by summing the scores assigned to the frequency of intake of each food item; higher scores indicate greater intake. A composite measure of fruit and vegetable intake was also estimated using an equation outlined by Block et al. This equation estimated food pyramid definitions of servings and incorporated a fruit and vegetable score (based on intake frequency) and sex (female=l):

Fruit and vegetable servings = [-0.23 + 0.37 (fruit/vegetable score) – 0.55 (sex)]

This composite measure of fruit and vegetable intake may differ from the sum of the individual servings because the measure is standardized by sex. The range of standardized fruit and vegetable servings was -0.8-8.7.
canadian discount drugs

Similarly, equations were used to estimate grams of daily total and saturated fat intake and incorporated a meat/snack score (based on intake frequency; higher scores indicate greater intake) and sex (female=l):

Total fat (grams) = [32.7 + 24 (meat/snack score) + 11.2 (sex)]

Saturated fat (grams) = [9.4 + 0.88 (meat/snack score)-3.5 (sex)]

The mean total fat and saturated fat grams ranged from 35.1-144.7 and 5.9^*8.1, respectively.

To evaluate correlates of dietary patterns, the different dietary domains stratified by sociodemographic, health and behavior variables were evaluated. T-tests were used to determine whether there were statistically significant differences in mean daily intake of servings of fruits and vegetables and the mean daily intake of grams of fat across these groups (comparing each group to the reference group). Based on this intake, the proportion of individuals meeting national guidelines for daily fruit, vegetable and fat intake was calculated. Multiple linear regression analyses were also conducted to determine the mean daily intake of fruit, vegetable and fat, stratified by correlates, after adjustment for age and sex. Results were similar after adjustment; therefore, crude results are presented.
kamagra oral jelly uk

Categories: Diet
Tags: , ,

© 2008 HIV/AIDS News & Information