Validation of a food-frequency questionnaire assessment of carotenoid and vitamin E intake using weighed food records and plasma biomarkers: the method of triads model.

Background: Reliability or validity studies are important for the evaluation of measurement error in dietary assessment methods. An approach to validation known as the method of triads uses triangulation techniques to calculate the validity coefficient of a food-frequency questionnaire (FFQ).

Objective: To assess the validity of an FFQ estimates of carotenoid and vitamin E intake against serum biomarker measurements and weighed food records (WFRs), by applying the method of triads. Design: The study population was a sub-sample of adult participants in a randomised controlled trial of b-carotene and sunscreen in the prevention of skin cancer. Dietary intake was assessed by a self-administered FFQ and a WFR. Nonfasting blood samples were collected and plasma analysed for five carotenoids (a-carotene, b-carotene, b-cryptoxanthin, lutein, lycopene) and vitamin E. Correlation coefficients were calculated between each of the dietary methods and the validity coefficient was calculated using
the method of triads. The 95% confidence intervals for the validity coefficients were estimated using bootstrap sampling.

Results: The validity coefficients of the FFQ were highest for a-carotene (0.85) and lycopene (0.62), followed by b-carotene (0.55) and total carotenoids (0.55), while the lowest validity coefficient was for lutein (0.19). The method of triads could not be used for b-cryptoxanthin and vitamin E, as one of the three underlying correlations was negative.

Conclusions: Results were similar to other studies of validity using biomarkers and the method of triads. For many dietary factors, the upper limit of the validity coefficients was less than 0.5 and therefore only strong relationships between dietary exposure and disease will be detected.

Reliability of a food frequency questionnaire to assess dietary antioxidant intake

OBJECTIVE: Epidemiologic evidence of a role for antioxidants in the prevention of chronic disease has been inconclusive, in part due to the difficulty of measuring past diets of free-living populations. The purpose of the current study was to examine the reliability of a 19-item, self-administered, semiquantitative, food frequency questionnaire to assess intake of the major dietary antioxidants. METHODS: Reliability was established by administering the food frequency questionnaire a second time by telephone. The subjects comprised 151 participants in the Melbourne Visual Impairment Project, a study of the distribution and determinants of eye disease in Melbourne residents aged 40 and over. RESULTS: Spearman correlation coefficients ranged from 0.39 for spinach to 0.76 for yoghurt, and all were highly significant (all p = 0.001). The reliability of the instrument was not influenced by gender, English speaking ability, or the number of days between the first and second administration of the questionnaire. CONCLUSION: In conclusion, we have shown this 19-item food frequency questionnaire to be highly reliable. It should be useful for anyone involved in the study of the relationship of dietary antioxidant intake to health outcomes in large populations where limitations of time and money prohibit the collection of more detailed dietary intake information.

Reproducibility of the online food4me food-frequency questionnaire for estimating dietary intakes across Europe

BACKGROUND: Accurate dietary assessment is key to understanding nutrition-related outcomes and is essential for estimating dietary change in nutrition-based interventions. OBJECTIVE: The objective of this study was to assess the pan-European reproducibility of the Food4Me food-frequency questionnaire (FFQ) in assessing the habitual diet of adults. METHODS: Participants from the Food4Me study, a 6-mo, Internet-based, randomized controlled trial of personalized nutrition conducted in the United Kingdom, Ireland, Spain, Netherlands, Germany, Greece, and Poland, were included. Screening and baseline data (both collected before commencement of the intervention) were used in the present analyses, and participants were included only if they completed FFQs at screening and at baseline within a 1-mo timeframe before the commencement of the intervention. Sociodemographic (e.g., sex and country) and lifestyle [e.g., body mass index (BMI, in kg/m(2)) and physical activity] characteristics were collected. Linear regression, correlation coefficients, concordance (percentage) in quartile classification, and Bland-Altman plots for daily intakes were used to assess reproducibility. RESULTS: In total, 567 participants (59% female), with a mean ± SD age of 38.7 ± 13.4 y and BMI of 25.4 ± 4.8, completed both FFQs within 1 mo (mean ± SD: 19.2 ± 6.2 d). Exact plus adjacent classification of total energy intake in participants was highest in Ireland (94%) and lowest in Poland (81%). Spearman correlation coefficients (ρ) in total energy intake between FFQs ranged from 0.50 for obese participants to 0.68 and 0.60 in normal-weight and overweight participants, respectively. Bland-Altman plots showed a mean difference between FFQs of 210 kcal/d, with the agreement deteriorating as energy intakes increased. There was little variation in reproducibility of total energy intakes between sex and age groups. CONCLUSIONS: The online Food4Me FFQ was shown to be reproducible across 7 European countries when administered within a 1-mo period to a large number of participants. The results support the utility of the online Food4Me FFQ as a reproducible tool across multiple European populations. This trial was registered at clinicaltrials.gov as NCT01530139.

Income differences in food consumption in the 1995 Australian national nutrition survey

Objective: To assess the relationships between an index of per capita income and the intake of a variety of individual foods as well as groups of food for men and women in different age groups. Design: Cross-sectional national survey of free-living men and women. Subjects: A sample of 5053 males and 5701 females aged 18 y and over who completed the Australian National Nutrition Survey 1995. Methods: Information about the frequency of consumption of 88 food items was obtained. On the basis of scores on the Food Frequency Questionnaire, regular and irregular consumers of single foods were identified. The relationships between regularity of consumption of individual foods and per capita income were analysed via contingency tables. Food variety scores were derived by assigning individual foods to conventional food group taxonomies, and then summing up the dichotomised intake scores for individual foods within each food group. Two-way ANOVA (income age group) were performed on the food variety scores for males and females, respectively. Results: Per capita income was extensively related to the reported consumption of individual foods and to total and food group variety indices. Generally, both men and women in low income households had less varied diets than those in higher-income households. However, several traditional foods were consumed less often by young high-income respondents, especially young women. Conclusions: Major income differentials in food variety occur in Australia but they are moderated by age and gender. Younger high-income women, in particular, appear to have rejected a number of traditional foods, possibly on the basis of health beliefs. The findings also suggest that data aggregation has marked effects on income and food consumption relationships.

The relationship between education and food consumption in the 1995 Australian national nutrition survey

Objective: To assess the relationship between education and the intake of a variety of individual foods, as well as groups of foods, for Australian men and women in different age groups.

Design: Cross-sectional national survey of free-living men and women. Subjects: A sample of 2501 men and 2739 women aged 18 years and over who completed the National Nutrition Survey (NNS) 1995.

Methods: Information about the frequency of consumption of 88 food items was obtained using a food-frequency questionnaire in a nation-wide nutrition survey. Irregular and regular consumers of foods were identified according to whether they consumed individual foods less than or more than once per month. The relationship between single foods and an index of education (no post-school qualifications, vocational, university) was analysed via contingency table chi-square statistics for men and women. Food group variety scores were derived by assigning individual foods to conventional food group taxonomies, and then summing the dichotomised intake scores for individual foods within each food group. Two-way analyses of variance (education by age groups) were performed on food variety scores for men and women, separately.

Results: While university-educated men and women consumed many individual foods more regularly than less-educated people, they were less likely to be regular consumers of several meat products. The relationship between education and food consumption was less apparent when individual food scores were aggregated into food group scores. University-educated men and women exhibited higher scores on total food group variety than the other educational groups.

Conclusions: Higher education is associated with the regular consumption of a wider variety of foods. Aggregation of individual food consumption indices into food variety scores may mask the apparent effects of educational background on food consumption.

Food intake patterns among Australian adolescents

Objectives: This study aimed 10 evaluate the food intake pattens of adolescents with respect to the Australian Guide to Healthy Eating, and to examine variations in food intake patterns by age, gender and region of residence.
Design: Cross-sectional online food survey administered through schools. Participants and setting: In 2004-2005,3841 secondary students in years seven (12-13 years) and nine (14-15 years) drawn from 37 secondary schools in Victoria, Australia completed an online food intake patterns survey.
Outcome measures: Food intake was measured by a Food Frequency Questionnaire (FFQ), and categorized according to the five basic food groups (fruit, vegetables, meat, daily, cereal) and the 'extra' food group as defined by the Australian Guide to Healthy Eating (AGHE). The foods groups were examined in the study population and compared across age, gender and region.
Results: Many adolescents in this sample reported food intakes that deviated substantially from recommendations of the AGHE. For example, two-thirds of participants failed to consume foods from the five recommended food groups daily; over a third reported eating fruit 'rarely or never'; and 22% reported eating fast foods every day. Food intakes were generally more in line with dietary guidelines among girls than boys.
Regional differences were less consistent, and there were few differences by age.
Conclusion: A significant proportion of adolescents have food intakes that fall short of the recommendations outlined in the Australian Guide to Healthy Eating. This highlights the need for public health initiatives to promote healthier food intake pattens among adolescents.

Associations between the home food environment and obesity-promoting eating behaviors in adolescence

Objective: This study examines relationships between multiple aspects of the home food environment and obesity-promoting characteristics of 12- to 13-year-old adolescents' diets, specifically frequency of consumption of high-energy fluids, sweet snacks, savory snacks, and take-out foods.

Research Methods: This was a cross-sectional study including 347 adolescents 12 to 13 years of age and their parents. Data were collected via self-completed surveys. The adolescents' diets were assessed using a Food Frequency Questionnaire derived from existing age-appropriate National Nutrition Survey data. An extensive range of domains within the home food environment were assessed. Bivariate linear regression analyses were run split by gender. Forced entry multiple linear regression analyses (adjusting for all variables significant in bivariate analyses as well as for maternal education) were also performed, stratified by the sex of the child.

Results: The influence of mothers, either as models for eating behaviors or as the providers of food, was pervasive. Mothers' intake of high-energy fluids (p = 0.003), sweet snacks (p = 0.010), savory snacks (p = 0.008), and take-out food (p = 0.007) was positively associated with boys' intake of all these foods. In addition, mothers' intake of high-energy fluids was positively associated with daughters' consumption of these drinks (p = 0.025). Furthermore, availability of unhealthy foods at home was positively associated with girls' sweet snack (p = 0.001), girls' savory snack (p < 0.001), boys' savory snack (p = 0.002), and, in the bivariate analyses, girls' high-energy fluid consumption (p = 0.002).

Discussion: This study of home food environment influences on adolescent diet highlights the pervasive influence of mothers in determining adolescents' obesity-promoting eating, providing direction for obesity prevention strategies and future research.

Takeaway food consumption, diet quality and abdominal obesity in young adults

Background - Takeaway food consumption is associated with a higher BMI and poorer diet quality in the USA but little is known about the association in Australians.
Objective - To examine if takeaway food consumption is associated with abdominal obesity and poorer diet quality in young Australian adults.
Design - A national sample of 1,277 men and 1,585 women aged 26-36 completed a self-administered questionnaire on demographic and lifestyle factors, a 127 item food frequency questionnaire, and usual frequency of fruit, vegetable and takeaway food consumption. Dietary intake was compared with the dietary recommendations of the Australian Guide to Healthy Eating. A pedometer was worn for seven days. Waist circumference was measured and moderate abdominal obesity was defined as ≥94 cm for men and ≥80 cm for women. Prevalence ratios (PR) were calculated using log binomial regression with eating takeaway food once a week or less as the reference group.
Outcomes - Consumption of takeaway food twice a week or more was reported by more men (37.9%) than women (17.7%). Participants eating takeaway food at least twice a week were less likely to meet the guidelines for vegetables (P<0.05 men and women), fruit (P<0.001 men and women), dairy (P<0.01 men and women), extra foods (P=0.001 men and women), breads and cereals (P<0.05 men only), lean meats and alternatives (P<0.05 women only) and overall met significantly fewer dietary guidelines (P<0.001 men and women) than participants eating takeaway less than twice per week. After adjusting for confounding variables (age, physical activity, TV viewing, and employment status) consuming takeaway food twice a week or more was associated with a 31% higher prevalence of moderate abdominal obesity in men (PR 1.31; 95% CI: 1.07, 1.61) and a 25% higher prevalence in women (PR 1.25; 95% CI: 1.04, 1.50).
Conclusion - Eating takeaway food twice a week or more was associated with poorer diet quality and a higher prevalence of moderate abdominal obesity in both young men and young women.

Takeaway food consumption and its associations with diet quality and abdominal obesity : a cross-sectional study of young adults

Background : Few studies have investigated the associations of takeaway food consumption with overall diet quality and abdominal obesity. Young adults are high consumers of takeaway food so we aimed to examine these associations in a national study of young Australian adults.

Methods : A national sample of 1,277 men and 1,585 women aged 26–36 completed a self-administered questionnaire on demographic and lifestyle factors, a 127 item food frequency questionnaire, usual daily frequency of fruit and vegetable consumption and usual weekly frequency of takeaway food consumption. Dietary intake was compared with the dietary recommendations from the Australian Guide to Healthy Eating. Waist circumference was measured for 1,065 men and 1,129 women. Moderate abdominal obesity was defined as ≥ 94 cm for men and ≥ 80 cm for women. Prevalence ratios (PR) were calculated using log binomial regression. Takeaway food consumption was dichotomised, with once a week or less as the reference group.

Results : Consumption of takeaway food twice a week or more was reported by more men (37.9%) than women (17.7%, P < 0.001). Compared with those eating takeaway once a week or less, men eating takeaway twice a week or more were significantly more likely to be single, younger, current smokers and spend more time watching TV and sitting, whereas women were more likely to be in the workforce and spend more time watching TV and sitting. Participants eating takeaway food at least twice a week were less likely (P < 0.05) to meet the dietary recommendation for vegetables, fruit, dairy, extra foods, breads and cereals (men only), lean meat and alternatives (women only) and overall met significantly fewer dietary recommendations (P < 0.001). After adjusting for confounding variables (age, leisure time physical activity, TV viewing and employment status), consuming takeaway food twice a week or more was associated with a 31% higher prevalence of moderate abdominal obesity in men (PR: 1.31; 95% CI: 1.07, 1.61) and a 25% higher prevalence in women (PR: 1.25; 95% CI: 1.04, 1.50).

Conclusion : Eating takeaway food twice a week or more was associated with poorer diet quality and a higher prevalence of moderate abdominal obesity in young men and women.

Family food environments as determinants of children's eating : implications for obesity prevention

The prevalence of childhood obesity is escalating rapidly and it considered to be a major public health problem. Diet is a recognised precursor of fatness, and current evidence supports the premise that in Westernised countries, the dietary intakes of children are likely to be important in obesity genesis. However, we have a relatively poor understanding of the environments in which a child’s eating is learnt and maintained. Much of the existing work in this area is based on small-scale or experimental studies, or has been derived from homogeneous populations within the USA. Despite these limitations, there is evidence that aspects of the child’s family environment are likely to be important in determining obesity risk in children. This thesis examines the impact of the family food environment on a child’s eating through two related studies. The first study, titled the Children and Family Eating (CAFÉ) study comprised three phases. Phase one involved qualitative interviews with 17 parents of 5-6 year-old children to explore parental perceptions regarding those factors in a child’s environment believed to influence the development of their child’s eating habits. These interviews were used to inform the development of quantitative measures of the family food environment. The second phase involved the development of a Food Frequency Questionnaire (FFQ) to assess dietary intake in 5-6 year-olds. The FFQ was informed by analysis of 1995 Australian National Nutrition Survey data. In the final phase the relationships between dietary intakes of 5-6 year-old children, and potential predictors of dietary intake were examined in a cross-sectional study of 560 families. Predictors included measures of: parental perceptions of the adequacy of their child’s diet; food availability and accessibility; child-feeding; the opportunities for parental modelling of food intake; a child’s television exposure; maternal Body Mass Index; and maternal education. Analysis of the CAFÉ data provides unique information regarding the relationships between a child’s family food environment and their food consumption. Models developed for a range of dietary outcomes considered to be predictive of increased risk for obesity, including total energy and fat intakes, vegetable variety, vegetable consumption, and high-energy (non-dairy) fluid consumption, explained between 11 and 20 percent of the variance in dietary intake. Two aspects of the family food environment, parental perception of a child’s dietary adequacy, and the total minutes of television viewed per day, were frequently found to be predictive of dietary outcomes likely to promote fatness in these children. The second study, titled the Parent Education and Support (PEAS) Feeding Intervention Study, was a prospective pre/post non-randomised intervention trial that assessed the impact of a feeding intervention to 240 first-time mothers of one-year-old children. This intervention focused on one aspect of the family food environment, child-feeding, which has been proposed as influential in the development of obesogenic eating behaviours. In this study, Maternal and Child Health Nurses (MCHNs), using a ‘Division of Responsibility’ model of feeding, taught parents to provide nutritious food at regular intervals and to let children decide if to eat and how much to eat. Thus parents were encourages to food their child without exerting pressure, or employing coercion or rewards (controlling behaviours). The aim was to influence parental attitudes and beliefs regarding child-feeding. Through the use of these feeding techniques, this intervention also aimed to increase the variety of fruits and vegetables a child consumed by teaching parents to persist with offering these foods, over the year of the intervention, in non-emotive environments. Fruits and vegetables were chosen in this intervention because they are likely to be protective in the development of obesity. Analysis of the PEAS data suggests that this low-level feeding intervention, delivered through existing Maternal and Child Health services, was modestly effective in changing parental attitudes and beliefs regarding the feeding of young children. Further, the validity of fruits offered to intervention group children increased. This thesis expands the existing knowledge base by providing a comprehensive analysis of the relative impact of aspects of the family environment on dietary intakes of 5-6 year-olds. Further, the analysis of a feeding intervention in first-time parents provides important insights regarding the potential to influence child-feeding and the impact this may have on the promotion of eating behaviours protective against obesity.

Factors influencing reported food intake

A study was conducted to investigate the possible influence of body size and composition, errors in portion size estimation, level of satisfaction with body size and degree of dietary restraint on the level of reported food intake obtained using both a food frequency questionnaire and weighed food record. The findings suggest that in dietary studies based on weighed food records, 'weight consciousness', as determined using a measure of dietary restraint, could be more important than influence of body size in determining the level of reported intake.

Food store choice among urban slum women is associated with consumption of energy-dense food

The aim of this study was to examine the associations of food store choice with food consumption among urban slum women. A cross-sectional survey was carried out among 188 urban slum women (19-50 years old) in Jakarta, Indonesia. A semiquantitative food frequency questionnaire was used to assess food consumption. Associations between food consumption and food store choice were tested by linear regression. This study found that frequencies of buying food from small shops (warung), street food vendors, and modern food stores were significantly associated with consumption of snacks, mixed dishes, and fruit respectively. In addition, buying food from traditional markets and small cafes (warung makan) was not significantly associated with particular types of food consumption. As modern food stores are rarely utilized by these women, small shops (warung) and street food vendors are likely to be important channels to improve slum dwellers’ diet.

How well do Australian women comply with dietary guidelines?

Objective: To investigate the proportion of middle-aged Australian women meeting national dietary recommendations and its variation according to selected sociodemographic and behavioural characteristics.

Design: This cross-sectional population-based study used a food-frequency questionnaire to investigate dietary patterns and compliance with 13 commonly promoted dietary guidelines among a cohort of middle-aged women participating in the Australian Longitudinal Study on Women's Health.

Setting: Nation-wide community-based survey.

Subjects: A total of 10 561 women aged 50–55 years at the time of the survey in 2001.

Results: Only about one-third of women complied with more than half of the guidelines, and only two women in the entire sample met all 13 guidelines examined. While guidelines for meat/fish/poultry/eggs/nuts/legumes and ‘extra’ foods (e.g. ice cream, chocolate, cakes, potatoes, pizza, hamburgers and wine) were met well, large percentages of women (68–88%) did not meet guidelines relating to the consumption of breads, cereal-based foods and dairy products, and intakes of total and saturated fat and iron. Women working in lower socio-economic status occupations, and women living alone or with people other than a partner and/or children, were at significantly increased risk of not meeting guidelines.

Conclusions: The present results indicate that a large proportion of middle-aged Australian women are not meeting dietary guidelines. Without substantial changes in their diets, and help in making these changes, current national guidelines appear unachievable for many women.

Dietary intake in post-menopausal women

Si has been suggested as an essential element, and may be important in optimal bone, skin and cardiovascular health. However, there are few estimates of dietary Si intakes in man, especially in a UK population. Following the development of a UK food composition database for Si, the aim of the present study was to investigate dietary intakes of Si amongst healthy women aged over 60 years and to identify important food sources of Si in their diet. Healthy, post-menopausal female subjects (>60 years of age; n 209) were recruited from the general population around Dundee, Scotland as part of an unrelated randomised controlled intervention study where dietary intake was assessed using a self-administered, semi-quantitative food-frequency questionnaire at five time-points over a 2-year period. Food composition data on the Si content of UK foods was used to determine the Si content of food items on the food-frequency questionnaire. Mean Si intake was 18·6 (sd 4·6) mg and did not vary significantly across the 2 years of investigation. Cereals provided the greatest amount of Si in the diet (about 30%), followed by fruit, beverages (hot, cold and alcoholic beverages combined) and vegetables; together these foods provided over 75% about Si intake. Si intakes in the UK appear consistent with those reported previously for elderly women in Western populations, but lower than those reported for younger women or for men.