What do babies eat? Evaluation of a food frequency questionnaire to assess the diets of infants aged 6 months

OBJECTIVE: To evaluate the relative validity of a food frequency questionnaire (FFQ) for assessing nutrient intakes in 6-month-old infants. DESIGN AND SETTING: The FFQ was developed to assess the diets of infants born to women in the Southampton Women's Survey (SWS), a population-based survey of young women and their offspring. The energy and nutrient intakes obtained from an interviewer-administered FFQ were compared with those obtained from a 4-day weighed diary. SUBJECTS AND METHODS: A sub-sample of 50 infants aged 6 months from the SWS had their diets assessed by both methods. The FFQ recorded the frequencies and amounts of milks, baby foods, regular foods and drinks consumed by the infants over the previous seven days. The diaries recorded the weights of all foods and drinks consumed by the infants on four separate days within 15 days following FFQ completion. RESULTS: Spearman rank correlation coefficients for intakes of energy, macronutrients and 18 micronutrients, determined by the two methods, ranged from r = 0.39 to 0.86; adjustment for energy intake tended to increase the correlation coefficients, range r(a) = 0.55 to 0.89. Bland-Altman statistics showed that mean differences between methods were in the range of -12.5% to +12.5% except for vitamin B12 (-18.9%). CONCLUSION: Although there were differences in absolute energy and nutrient intakes between methods, Spearman rank correlation coefficients indicated reasonable agreement in the ranking of intakes. The interviewer-administered FFQ is a useful tool for assessing energy and nutrient intakes of healthy infants aged about 6 months.     

Carotenoid intakes, assessed by food-frequency questionnaires (FFQs), are associated with serum carotenoid concentrations in the Jackson Heart Study: validation of the Jackson Heart Study Delta NIRI Adult FFQs

OBJECTIVES: Intake and status of carotenoids have been associated with chronic disease. The objectives of this study were to examine the association between carotenoid intakes as measured by two regional food-frequency questionnaires (FFQs) and their corresponding measures in serum, and to report on dietary food sources of carotenoids in Jackson Heart Study (JHS) participants. DESIGN: Cross-sectional analysis of data for 402 African American men and women participating in the Diet and Physical Activity Sub-Study (DPASS) of the JHS. RESULTS: Mean serum carotenoid concentrations and intakes in this population were comparable to those reported for the general US population. After adjustment for covariates, correlations between serum and dietary measures of each carotenoid, for the average of the recalls (deattenuated), the short FFQ and the long FFQ, respectively, were: 035 and 0-carotene; 026 and 0-carotene; 017 and 0-carotene; 034 and 0-cryptoxanthin; 015 and 037, 014 for lycopene. Major dietary sources of -carotene and lutein plus zeaxanthin, mustard, turnip and collard greens; of beta-cryptoxanthin, orange juice; and of lycopene, tomato juice. CONCLUSIONS: On average, carotenoid intakes and serum concentrations are not lower in this southern African American population than the general US population. The two regional FFQs developed for a southern US population and used as dietary assessment tools in the JHS appear to provide reasonably valid information for most of these carotenoids.     

Development and simulated validation of a food-frequency questionnaire for the Colombian population

OBJECTIVE: To develop a food-frequency questionnaire (FFQ) useful for ranking of nutrient intakes. DESIGN: Subjects consuming their regular diet completed 7 days of weighed intake registry (7-WIR). Foods for the FFQ were selected by stepwise multiple regression. The FFQ was then completed for each subject using data on individual food consumption from the 7-WIR. The correlation and agreement between the extrapolated FFQ and the 7-WIR data were assessed using Spearman's rank correlation coefficients (rS) and Bland and Altman's limits of agreement (LOA). SETTING: Bucaramanga, Colombia. SUBJECTS: We studied 97 randomly selected 20-40-year-old subjects. RESULTS: Sixty foods were selected for the FFQ. The 7-WIR and the extrapolated FFQ intake estimates correlated well. rS was 0.58 for energy, 0.53 for carbohydrate, 0.50 for total fat, and 0.48 for protein. For micronutrients, rS varied from 0.46 (manganese) to 0.71 (vitamin B12). FFQ average intake estimates were 83%, 80%, 86.2% and 86.4% of 7-WIR estimates for energy, carbohydrate, total fat and protein, respectively. LOA for these nutrients ranged between 45% and 165%. FFQ micronutrient intakes were on average 96% (median) of those from the 7-WIR, and the median lower and upper LOA were 50% and 203%. However, there was no indication that the degree of agreement varied with the level of intake. CONCLUSIONS: According to our simulated validation, this FFQ may be useful to rank subjects by nutrient intake. Its validity against standard independent measurements and its applicability to other subsets of the Colombian population should be carefully considered.     

Comparative validation of standard, picture-sort and meal-based food-frequency questionnaires adapted for an elderly population of low socio-economic status

OBJECTIVE: To compare the validity of a modified Block food-frequency questionnaire (FFQ), a picture-sort administration of the FFQ (PSFFQ) and a meal pattern-based questionnaire (MPQ) in a multi-ethnic population of low socio-economic status (SES). DESIGN: Participants completed six 24-hour dietary recalls (24HR) over six months; the FFQ, PSFFQ and MPQ were completed in random order in the subsequent month. Instruments were interviewer-administered. The PSFFQ and MPQ were developed in formative research concerning difficulties for older adults in responding to standard food-frequency instruments. SETTING: Rural North Carolina, USA.Subjects One hundred and twenty-two African American, Native American and white adults aged > or = 65 years, with approximately one-third in each ethnic group. Inclusion criteria included education < or = 12 years and income < or = 150% of national poverty level or Medicaid recipient. RESULTS: Comparing median intakes from the average of the 24HR with the three diet assessment instruments, the MPQ tended to overestimate intakes compared with the FFQ and PSFFQ. Correlations among nutrients obtained by the 24HR and the other three instruments were generally statistically significant and positive. Across nutrients, the PSFFQ was most highly correlated with the 24HR for women, while the FFQ was most highly correlated with the 24HR for men. CONCLUSIONS: Dietary assessments using 24HR and FFQ were similar to results reported elsewhere, although correlations between 24HR and FFQ were somewhat lower. Interviewer-administered dietary assessments should be used with caution to evaluate dietary intake among older adults with low SES. Gender differences and the lower correlations should be investigated more thoroughly to assist in choosing dietary assessment instruments for this population.     

Validation of a food-frequency questionnaire for use in pregnancy

OBJECTIVES: As a part of an ongoing project to develop a nutritional screening tool, we evaluated the performance of a semi-quantitative food-frequency questionnaire (FFQ) in terms of validity in a Sheffield Caucasian pregnant population using two different statistical approaches--the correlation coefficient and the limits of agreement (LOA). The FFQ was designed specifically for pregnant women and previously used in a large-scale study. DESIGN: A validation study. SETTING: A community-based field study of a general population of pregnant women booked for their first antenatal appointment at the Jessop Wing, Royal Hallamshire Hospital, Sheffield, UK. SUBJECTS: One hundred and twenty-three women of different socio-economic status, aged between 17 and 43 years, provided complete dietary data. RESULTS: The validity of the FFQ was tested against a series of two 24-hour recalls. As expected, the intakes of all examined nutrients, except for iodine, carotene, vitamin E, biotin, vitamin C and alcohol, were higher when determined by the FFQ than when determined by 24-hour recall. Pearson's correlation coefficient between the two methods ranged from 0.19 (added sugar, zinc) to 0.47 (Englyst fibre). The LOA were broader for some of the nutrients, e.g. protein, Southgate fibre and alcohol, and an increasing lack of agreement between the two methods was identified with higher dietary intakes. CONCLUSIONS: The FFQ gave useful estimates of the nutrient intakes of Caucasian pregnant women and appears to be a valid tool for categorising pregnant women according to dietary intake. The FFQ performed well for most nutrients and had acceptable agreement with the 24-hour recall.     

Validation of a semi-quantitative food-frequency questionnaire used among 2-year-old Norwegian children

OBJECTIVES: An adequate diet is of profound importance in infancy and early childhood. To ensure an optimal diet, knowledge about actual intake must be obtained. The aims of this study were to assess the validity of a semi-quantitative food-frequency questionnaire (SFFQ) applied in a large nation-wide survey among 2-year-old children and to examine the validity of the SFFQ in relation to different background parameters. DESIGN: The SFFQ was administered to the parents close to the child's second birthday, and one to two weeks later they started to weigh and record the child's diet for 7 days. SUBJECTS: One-hundred and eighty-seven families with a 2-year-old child completed both methods. RESULTS: There were no differences between the intakes of protein, saturated fatty acids, total carbohydrates and calcium estimated from the two methods. The average intake of all micronutrients, except for calcium, was overestimated by the SFFQ. Bland-Altman plots showed a systematic increase in difference between the two methods with increasing intake for most nutrients. Spearman correlation coefficients between methods for nutrient intakes ranged from 0.26 to 0.50, the median correlation was 0.38. The correlations increased when estimates were adjusted for energy intake, the median correlation being 0.52. Differences in observed validity were found according to the number of siblings.CONCLUSIONS: This study indicates that the SFFQ may be a valuable tool for measuring average intakes of energy, macronutrients and several food items among a 2-year-old population in Norway. The ability of the questionnaire to rank children according to intakes of nutrients and food items was rather low.     

Comparison of children's diets as reported by the child via the Youth/Adolescent Questionnaire and the parent via the Willett food-frequency questionnaire

OBJECTIVE: A comparison of a parent-completed Willett food-frequency questionnaire (FFQ) and a self-completed Youth/Adolescent Questionnaire (YAQ) has not yet been conducted. SETTING: In the Diabetes Autoimmunity Study in the Young (DAISY), parents report their child's diet on the FFQ annually from birth until age 10 years, when the child begins to report their own diet using the YAQ. SUBJECTS: To determine the comparability of these collection methods, 89 children aged 10-17 years and their parents completed the YAQ and FFQ, respectively, for the child's previous year's diet. DESIGN: We compared reported intakes for energy, the macronutrients and a variety of micronutrients of interest to the DAISY study. RESULTS: Bland-Altman plots of energy-adjusted differences between questionnaire responses against their means suggested that the two collection methods gave similar results. The average Spearman correlation coefficient of all energy-adjusted nutrient intakes was 0.50, and did not differ significantly by gender (males, r=0.48; females, r=0.46) or age (10-11 years, r=0.49; 12-17 years, r=0.51). While correlated, the nutrient values from the FFQ were higher than the nutrient values from the YAQ. CONCLUSIONS: While reported nutrient intakes are correlated, an indicator variable defining which survey method a nutrient was collected with should be included in any longitudinal data analyses examining nutrient intakes collected with the YAQ and the FFQ as independent predictors of a disease outcome.     

Performance of a short tool to assess dietary intakes of fruits and vegetables, percentage energy from fat and fibre

OBJECTIVES: We describe the methods used to develop and score a 17-item 'screener' designed to estimate intake of fruit and vegetables, percentage energy from fat and fibre. The ability of this screener and a food-frequency questionnaire (FFQ) to measure these exposures is evaluated. DESIGN: Using US national food consumption data, stepwise multiple regression was used to identify the foods to be included on the instrument; multiple regression analysis was used to develop scoring algorithms. The performance of the screener was evaluated in three different studies. Estimates of intakes measured by the screener and the FFQ were compared with true usual intake based on a measurement error model. SETTING: US adult population. SUBJECTS: For development of instrument, n=9323 adults. For testing of instrument, adult men and women in three studies completing multiple 24-hour dietary recalls, FFQ and screeners, n=484, 462 and 416, respectively. RESULTS: Median recalled intakes for examined exposures were generally estimated closely by the screener. In the various validation studies, the correlations between screener estimates and estimated true intake were 0.5-0.8. In general, the performances of the screener and the full FFQ were similar; estimates of attenuation were lower for screeners than for full FFQs. CONCLUSIONS: When coupled with appropriate reference data, the screener approach described may yield useful estimates of intake, for both surveillance and epidemiological purposes.     

Do aggregates of multiple questions better capture overall fish consumption than summary questions?

OBJECTIVE: To compare intake estimates, validity and reliability of two summary questions to measure fish consumption with information from a detailed semi-quantitative food-frequency questionnaire (FFQ) on fish consumption. DESIGN: Population-based, cross-sectional study. Participants completed an FFQ and provided blood samples for erythrocyte membrane eicosapentaenoic acid (EPA) analysis. Aggregate measures of consumption of fresh/frozen/canned fish (fresh fish) and smoked/salted/dried fish (preserved fish) were generated from the FFQ and were compared with responses to the summary questions regarding intakes of similar items. Both methods were tested for validity, using correlation and linear regression techniques with EPA, and retest reliability. SETTING: Perth metropolitan area, Western Australia. SUBJECTS: One hundred and nine healthy volunteers of both sexes, aged 21-75 years. RESULTS: The summary fresh fish measure underestimated frequency and grams per week given by the aggregate question by about 50%, while estimates from the summary preserved fish measure were approximately three times that of the aggregate measure. Multiple linear regression analysis suggested that the aggregates accounted for more of the variation in EPA levels, but the difference was minimal. Intra-class correlations confirmed that both methods were reliable. CONCLUSIONS: Our study indicates that extensive questioning results in different absolute intakes of fish compared with brief questioning, but does not add any information if ranking individuals according to overall consumption of fish.     

The effect of personal characteristics on the validity of nutrient intake estimates using a food-frequency questionnaire

OBJECTIVE: To assess validity of the Nambour food-frequency questionnaire (FFQ) relative to weighed food records (WFRs), and the extent to which selected demographic, anthropometric and social characteristics explain differences between the two dietary methods.DESIGN: Inter-method validity study; 129-item FFQ vs. 12 days of WFR over 12 months.SETTING: Community-based Nambour Skin Cancer Prevention Trial.SUBJECTS: One hundred and fifteen of 168 randomly selected participants in the trial (68% acceptance rate) aged 25-75 years.RESULTS: Spearman correlations between intakes from the two methods ranged from 0.18 to 0.71 for energy-adjusted values. Differences between FFQ and WFR regressed on personal characteristics were significantly associated with at least one characteristic for 16 of the 21 nutrients. Sex was significantly associated with differences for nine nutrients; body mass index (BMI), presence of any medical condition and age were each significantly associated with differences for three to six nutrients; use of dietary supplements and occupation were associated with differences for one nutrient each. There was no consistency in the direction of the significant associations. Regression models explained from 7% (riboflavin) to 27% (saturated fat) of variation in differences in intakes.CONCLUSIONS: The relative validity of FFQ estimates for many nutrients is quite different for males than for females. Age, BMI, medical condition and level of intake were also associated with relative validity for some nutrients, resulting in the need to adjust intakes estimates for these in modelling diet-disease relationships. Estimates for cholesterol, beta-carotene equivalents, retinol equivalents, thiamine, riboflavin and calcium would not benefit from this.     

Validation of a food-frequency questionnaire for measurement of nutrient intake in a dietary intervention study

OBJECTIVE: To validate a 171-item semi-quantitative food-frequency questionnaire (FFQ) for measurement of nutrient intake in an intervention trial based on walnut supplementation. DESIGN AND SETTING: Free-living adults from Southern California were randomly assigned to either an intervention (walnut-supplemented) or a control diet. The prescribed 6-month intervention was > or = 28 g of walnuts per day for the walnut-supplemented group and > or = 2 g of walnuts per day for the control group. Participants provided at least six 24-hour dietary recalls and completed a self-administered FFQ. SUBJECTS: Eighty-seven adults aged 30-72 years (48 females, 39 males). RESULTS: Our findings from validation (by correlation with six diet recall measures) of the measurement of 32 nutrients by the FFQ are as follows. We found significant positive correlations (corrected for measurement error) between the FFQ and diet recalls for total energy (r = 0.34), total carbohydrate (r = 0.42), vegetable protein (r = 0.43), total fat (r = 0.51), polyunsaturated fat (r = 0.77), total fibre (r = 0.60), linoleic acid (r = 0.78) and alpha-linolenic acid (r = 0.79) - the last nutrient being an excellent nutrient biomarker of the intervention (walnut supplementation). Significant positive correlations were also found for vitamin C (r = 0.96) and certain minerals (r = 0.46-0.80 for calcium, phosphorus, magnesium, iron and potassium). Uncorrected correlations were also high (r>0.40) for retinol, beta-carotene, folate and alcohol. Both diet recalls and FFQ showed a similar significant difference in alpha-linolenic acid content between the walnut-supplemented and control diets. CONCLUSIONS: The FFQ demonstrated good relative validity in the estimation intake of some of the major nutrients in a dietary intervention trial and was a particularly valid estimate of an important nutrient biomarker of walnut supplementation.     

Issues in assessing the validity of nutrient data obtained from a food-frequency questionnaire: folate and vitamin B12 examples

OBJECTIVE: To compare methods used to assess the validity of nutrient intake data obtained from a food-frequency questionnaire (FFQ), using folate and vitamin B12 as nutrient examples. DESIGN: Cross-sectional sample from a population cohort. SETTING: Two postcode areas west of Sydney, Australia. SUBJECTS: In total, 2895 people aged 49 years and older provided dietary data using a semi-quantitative FFQ (79% of 3654 subjects examined). The validity of the FFQ was assessed against three 4-day weighed food records (WFRs) completed by 78 people (mean age 70 years). RESULTS: Folate and vitamin B12 validity data were assessed using different methods. The Spearman ranked correlations (energy-adjusted) were 0.66 for folate and 0.38 for vitamin B12. Using the Bland-Altman method, following loge transformation, no linear trend existed between the differences and means for folate and vitamin B12. Large differences existed between the FFQ and WFR in individual cases, particularly for vitamin B12. Finally, data were divided into quintile categories for the test and reference method: 79% classified folate within one quintile, 65% classified vitamin B12 within one quintile; there was no gross misclassification for folate and only 3% misclassification for vitamin B12. CONCLUSIONS: Different methods of analysis provided different information about the validity of the FFQ. Correlation coefficients should not be used alone to assess the validity of nutrient data, but should be used in conjunction with Bland-Altman analyses. Depending on the use of the data, additional assessment of classification categories is recommended. This worked example demonstrates that absolute intakes of folate and vitamin B12 should be used with caution.     

Comparing methods for handling missing values in food-frequency questionnaires and proposing k nearest neighbours imputation: effects on dietary intake in the Norwegian Women and Cancer study (NOWAC)

OBJECTIVE: To investigate item non-response in a postal food-frequency questionnaire (FFQ), and to assess the effect of substituting/imputing missing values on dietary intake levels in the Norwegian Women and Cancer study (NOWAC). We have adapted and probably for the first time applied k nearest neighbours (KNN) imputation to FFQ data. DESIGN: Data from a recent reproducibility study were used. The FFQ was mailed twice (test-retest) about 3 months apart to the same subjects. Missing responses in the test FFQ were imputed using the null value (frequencies = null, amount = smallest), the sample mode, the sample median, KNN, and retest values. SETTING: A methodological substudy of NOWAC, a national population-based cohort. SUBJECTS: A random sample of 2000 women aged 46-75 years was drawn from the cohort in 2002 (response 75%). The imputation methods were compared for 1430 women who completed at least 50% of the test FFQ. RESULTS: We imputed 16% missing values in the overall test data matrix. Compared to null value imputation, the largest differences in estimated dietary intake were seen for KNN, and for food items with a high proportion of missing. Imputation with retest values increased total energy intake, indicating that not all missing values are caused by respondents failing to specify no consumption, and that null value imputation may lead to underestimation and misclassification. CONCLUSION: Missing values in FFQs present a methodological challenge. We encourage the application and evaluation of newer imputation methods, including KNN, which may reduce imputation errors and give more accurate intake estimates.     

Performance of a food-frequency questionnaire in the US NIH-AARP (National Institutes of Health-American Association of Retired Persons) Diet and Health Study

OBJECTIVE: We evaluated the performance of the food-frequency questionnaire (FFQ) administered to participants in the US NIH-AARP (National Institutes of Health-American Association of Retired Persons) Diet and Health Study, a cohort of 566 404 persons living in the USA and aged 50-71 years at baseline in 1995. DESIGN: The 124-item FFQ was evaluated within a measurement error model using two non-consecutive 24-hour dietary recalls (24HRs) as the reference. SETTING: Participants were from six states (California, Florida, Pennsylvania, New Jersey, North Carolina and Louisiana) and two metropolitan areas (Atlanta, Georgia and Detroit, Michigan). SUBJECTS: A subgroup of the cohort consisting of 2053 individuals. RESULTS: For the 26 nutrient constituents examined, estimated correlations with true intake (not energy-adjusted) ranged from 0.22 to 0.67, and attenuation factors ranged from 0.15 to 0.49. When adjusted for reported energy intake, performance improved; estimated correlations with true intake ranged from 0.36 to 0.76, and attenuation factors ranged from 0.24 to 0.68. These results compare favourably with those from other large prospective studies. However, previous biomarker-based studies suggest that, due to correlation of errors in FFQs and self-report reference instruments such as the 24HR, the correlations and attenuation factors observed in most calibration studies, including ours, tend to overestimate FFQ performance. CONCLUSION: The performance of the FFQ in the NIH-AARP Diet and Health Study, in conjunction with the study's large sample size and wide range of dietary intake, is likely to allow detection of moderate (> or =1.8) relative risks between many energy-adjusted nutrients and common cancers.     

Effect of the choice of food composition table on nutrient estimates: a comparison between the British and American (Chilean) tables

Objective: To determine the level of agreement between the American (Chilean) and British food composition tables in estimating intakes of macronutrients and antioxidants. DESIGN, SETTING AND SUBJECTS: Information based on a food-frequency questionnaire with emphasis on antioxidants was collected from 95 Chileans aged 24-28 years. Nutritional composition was analysed using the British table of food composition and the American table of food composition modified by Chilean food items. Mean differences and limits of agreement (LOAs) of estimated intake were assessed. RESULTS: Mean differences between the two tables of food composition ranged from 5.3% to 8.9% higher estimates when using the American (Chilean) table for macronutrients. For micronutrients, a bias towards a higher mean was observed for vitamin E, iron and magnesium when the American (Chilean) table was used, but the opposite was observed for vitamin A and selenium. The intra-class correlation coefficient (ICC) ranged from 0.86 (95% confidence interval (CI) 0.81-0.91) to 0.998 (95% CI 0.995-1.00), indicating high to excellent agreement. LOAs for macronutrients and vitamins A and C were satisfactory, as they were sufficiently narrow. There was more uncertainty for other micronutrients. CONCLUSION: The American table gives relative overestimates of macronutrients in comparison to the British table, but the relative biases for micronutrients are inconsistent. Estimates of agreement between the two food composition tables provide reassurance that results are interchangeable for the majority of nutrients.     

A regional food-frequency questionnaire for the US Mississippi Delta

OBJECTIVE: To describe food sources of nutrient intake for white and African American adults in the Lower Mississippi Delta (LMD), and their use in the development of a regional food-frequency questionnaire (FFQ) based on an earlier version of the National Cancer Institute's Health Habits and History Questionnaire. DESIGN: We ranked food sources of energy, macronutrients, vitamins and minerals, and examined portion size distributions for 842 white and 857 African American residents aged 19 years and older, using 24-hour dietary intake recall data from a telephone survey of 36 LMD counties. These values were used to develop a regional FFQ, which was then field-tested with 100 subjects and revised to improve interpretability. SETTING: The LMD region of the USA. SUBJECTS: White and African American adult residents of the LMD. RESULTS: LMD African Americans obtained more of their energy and nutrient intakes from poultry, processed meat, salty snacks, fruit drinks, pork and cornbread; and less from milk, alcohol, legumes, salad dressing, butter/margarine and sweetened tea than did white residents. Regional foods not on nationally used FFQs included grits, turnip greens, okra, ham hocks, chitterlings, crawfish, catfish, cracklings, jambalaya, potato logs, chicken and dumplings, and sweet potato pie. Based on responses during field-testing, the questionnaire was also designed to add four portion sizes for each food item, presented as questions, rather than in grid format. CONCLUSIONS: Regional food use patterns differ from national patterns and furthermore differ between African American and white adults in the LMD. The resulting Delta NIRI FFQ for Adults should contribute to improved assessment of usual intake for use in studies of diet and health in this region.     

The UK Women's Cohort Study: comparison of vegetarians, fish-eaters and meat-eaters

BACKGROUND: This paper describes the development of the UK Women's Cohort Study and presents cohort baseline characteristics. METHODS: In total, 35,372 women, aged 35-69 years at recruitment, were selected to ensure a wide range of dietary intakes. Diet was assessed by a 217-item food-frequency questionnaire (FFQ). Detailed lifestyle information was collected by postal questionnaire. Vegetarians, fish-eaters and meat-eaters were compared. RESULTS: The cohort women are mainly white, well-educated, middle-class and married with children. They are health-conscious with only 11% current smokers and 58% taking dietary supplements. Twenty-eight per cent of subjects self-report as being vegetarian and 1% as vegan. However, only 18% are defined as 'vegetarian' from the FFQ. Fat provides 32% of energy; vitamin and mineral intakes are high, with a broad range of intakes. Meat-eaters are older, with a higher body mass index (BMI) and the lowest intakes of carbohydrate, fibre, vitamin C, folate, iron and calcium. Other fish-eaters are similar to vegetarians. Vegetarians have the lowest intakes of protein, fat and saturated fat. Oily fish-eaters have the lowest BMI; are the least likely to smoke or use full-fat milk; and are the most likely to use dietary supplements and consume the most fruit and vegetables. Oily fish-eaters have the highest total energy intake and vegetarians the lowest. Semi-skimmed milk, bread, potatoes, wine, bananas and muesli are important contributors to energy for all groups. CONCLUSION: A large cohort of middle-aged women has been created encompassing a wide range of different eating patterns, including diets currently of interest to research into protection against cancer and coronary heart disease. Participants will be followed up to study the effects of different food and nutrient intakes on long-term health outcomes.     

Urine flavonoids and plasma carotenoids in the validation of fruit, vegetable and tea intake during pregnancy in the Norwegian Mother and Child Cohort Study (MoBa)

OBJECTIVE: To validate a new food-frequency questionnaire (FFQ) for measuring the intake of fruit, vegetables and tea reported by women participating in the Norwegian Mother and Child Cohort Study (MoBa). DESIGN: Intake of fruits, vegetables and tea estimated by the FFQ was compared with urinary flavonoid excretion, plasma carotenoid concentration and intake measured by a 4-day weighed food diary (FD). The triangular method was applied to calculate FFQ validity coefficients using two independent biomarkers. SETTING AND SUBJECTS: One hundred and nineteen women participating in MoBa. RESULTS: The FFQ estimate of fruit intake was significantly correlated with urine phloretin (r = 0.33), citrus fruit/juice with urine hesperetin (r = 0.44), cooked vegetables with plasma alpha-carotene (r = 0.37), and tea with urine kaempferol (r = 0.41) (P < 0.01 for all). On average, 60% of the participants fell into the same or adjacent quintiles when classified by FFQ and biomarkers. Significant correlations between the FFQ and FD were found for fruit (r = 0.39), vegetables (r = 0.34), juices (r = 0.50) and tea (r = 0.53). The FFQ validity coefficient was 0.65 for citrus fruit/juice and 0.59 for cooked vegetables as calculated by the triangular method. CONCLUSIONS: The validation study shows that the MoBa FFQ can be used to estimate fruit, juice, vegetable and tea intake in pregnant Norwegian women, and to rank individuals within the distribution.     

Frequency of soy food consumption and serum isoflavone concentrations among Chinese women in Shanghai

OBJECTIVE: The food-frequency questionnaire (FFQ) can be an efficient tool to evaluate dietary intake in large, population-based studies, especially for specific foods. The objective of this study was to validate the assessment of soy and isoflavone (daidzein and genistein) intakes, measured by an FFQ, by comparing intakes with serum isoflavone concentrations. DESIGN AND SETTING: Soy and isoflavone intakes and serum isoflavone concentrations were determined as part of a case-control study of dietary factors and risks of benign breast disease and breast cancer. The FFQ, administered during an in-person interview, included six soy-specific line items. Blood was drawn within one week of FFQ completion. SUBJECTS: In total, 1823 women living in Shanghai, People's Republic of China. RESULTS: In this population, soybean milk, fresh bean curd and other bean foods were eaten once per week, and fermented bean curd, fried bean curd puff and soybeans were eaten less than once per week. A significant linear trend (P<0.01) in serum isoflavone concentrations across increasing categories of soy and isoflavone intakes was observed, indicating that soy and isoflavone intakes, measured by the FFQ, well distinguished serum isoflavone concentrations. Linear trends were also observed in both case and control groups in stratified analyses, suggesting little differential bias by case-control status. CONCLUSIONS: The results suggest that the FFQ provides a useful marker of soy food consumption and isoflavone exposure in this population.     

Validity of a food-frequency questionnaire for elderly men in southeast China

OBJECTIVE: To investigate the internal validity of a food-frequency questionnaire (FFQ) developed for a case-control study of prostate cancer in southeast China. DESIGN: A comprehensive questionnaire comprising a quantitative FFQ and a short food habit questionnaire (SFHQ) was developed and modified from previous cancer and nutritional studies. The Goldberg formula (ratio of energy intake (EI) to basal metabolic rate (BMR), EI/BMR) was used to assess the validity of the FFQ by making comparisons with physical activity levels. Physical activity levels were measured by the estimated total metabolic equivalents (MET) and the ratio of energy expenditure (EE) to BMR (EE/BMR). Correlation analyses were undertaken to compare the SFHQ variables with those of the quantitative FFQ. SETTING: Hangzhou, Zhejiang Province, China. SUBJECTS: A total of 404 men over 45 years old with or without prostate cancer were recruited from eight hospitals. RESULTS: The partial correlation coefficients, controlling for age and family history of prostate cancer, were moderate to high (P<0.05) for preserved foods intake, fat consumption and tea drinking variables between the SFHQ and the quantitative FFQ. The average EI/BMR was 1.72, with 76% of subjects exceeding the Goldberg cut-off value of 1.35. Apart from weight, BMI, EE/BMR and MET, there were no significant differences in characteristics between low (<1.35) and normal EI/BMR groups. CONCLUSIONS: The FFQ is demonstrated to be a valid instrument to measure energy and food intake for elderly men in southeast China.