Retrospectively Estimating Energy Intake and Misreporting From a Qualitative Food Frequency Questionnaire: An Example Using Australian Cohort and National Survey Data.
ABSTRACT: Qualitative food frequency questionnaires (Q-FFQ) omit portion size information from dietary assessment. This restricts researchers to consumption frequency data, limiting investigations of dietary composition (i.e., energy-adjusted intakes) and misreporting. To support such researchers, we provide an instructive example of Q-FFQ energy intake estimation that derives typical portion size information from a reference survey population and evaluates misreporting. A sample of 1,919 Childhood Determinants of Adult Health Study (CDAH) participants aged 26-36 years completed a 127-item Q-FFQ. We assumed sex-specific portion sizes for Q-FFQ items using 24-h dietary recall data from the 2011-2012 Australian National Nutrition and Physical Activity Survey (NNPAS) and compiled energy density values primarily using the Australian Food Composition Database. Total energy intake estimation was daily equivalent frequency × portion size (g) × energy density (kJ/g) for each Q-FFQ item, summed. We benchmarked energy intake estimates against a weighted sample of age-matched NNPAS respondents (n = 1,383). Median (interquartile range) energy intake was 9,400 (7,580-11,969) kJ/day in CDAH and 9,055 (6,916-11,825) kJ/day in weighted NNPAS. Median energy intake to basal metabolic rate ratios were 1.43 (1.15-1.78) in CDAH and 1.35 (1.03-1.74) in weighted NNPAS, indicating notable underreporting in both samples, with increased levels of underreporting among the overweight and obese. Using the Goldberg and predicted total energy expenditure methods for classifying misreporting, 65 and 41% of CDAH participants had acceptable/plausible energy intake estimates, respectively. Excluding suspected CDAH misreporters improved the plausibility of energy intake estimates, concordant with expected body weight associations. This process can assist researchers wanting an estimate of energy intake from a Q-FFQ and to evaluate misreporting, broadening the scope of diet-disease investigations that depend on consumption frequency data.
Project description:NOVA is a food classification system that categorises food items into one of four categories according to the extent and purpose of their processing: minimally processed food (MPF), processed culinary ingredient (PCI), processed food (PF), or ultra-processed food (UPF). The aim of this study was to determine the relative validity and reproducibility of a food frequency questionnaire (EAT5 FFQ) for measuring daily energy intake (EI kJ) and percentage of daily energy intake (EI%) from each NOVA group in New Zealand children. One hundred parents of five year old children completed the 123 item EAT5 FFQ on two occasions four weeks apart. A 3 day weighed diet record (WDR) was completed on non-consecutive randomly assigned days between FFQ appointments. The FFQ overestimated EI (both as kJ and %) from MPF and UPF, and underestimated intakes from PCI and PF, compared with the WDR. Bland-Altman plots indicated reasonably consistent agreement between FFQ and WDR for MPF and UPF but not PCI or PF. Correlation coefficients between the FFQ and WDR were acceptable for EI (%) for MPF (<i>r</i> = 0.31) and UPF (<i>r</i> = 0.30). The FFQ differentiated between the highest and lowest quartiles for EI (%) from MPF and UPF foods (<i>p</i>-values for the trends were 0.006 and 0.009 respectively), and for EI (kJ) from UPF foods (<i>p</i>-value for trend 0.003). Bland-Altman plots indicated consistent agreement between repeat administrations of FFQ for MPF and UPF only, while intra-class correlations suggested good reproducibility for EI (kJ and %) for all four NOVA categories (range 0.51-0.76). The EAT5 FFQ has acceptable relative validity for ranking EI (%) from MPF and UPF. It has good reproducibility for measuring EI from all four NOVA categories, in young children.
Project description:BACKGROUND/OBJECTIVES:Adults often misreport dietary intake; the magnitude varies by the methods used to assess diet and classify participants. The objective was to quantify the accuracy of the Goldberg method for categorizing misreporters on a food frequency questionnaire (FFQ) and two 24-h recalls (24HRs). SUBJECTS/METHODS:We compared the Goldberg method, which uses an equation to predict total energy expenditure (TEE), with a criterion method that uses doubly labeled water (DLW), in a study of 451 men and women. Underreporting was classified using recommended cut points and calculated values. Sensitivity and specificity, positive predictive value (PPV) and negative predictive value and the area under the receiver operating characteristic curve (AUC) were calculated. Predictive models of underreporting were contrasted for the Goldberg and DLW methods. RESULTS:AUCs were 0.974 and 0.972 on the FFQ, and 0.961 and 0.938 on the 24HR for men and women, respectively. The sensitivity of the Goldberg method was higher for the FFQ (92%) than the 24HR (50%); specificity was higher for the 24HR (99%) than the FFQ (88%); PPV was high for the 24HR (92%) and FFQ (88%). Simulation studies indicate attenuation in odds ratio estimates and reduction of power in predictive models. CONCLUSIONS:Although use of the Goldberg method may lead to bias and reduction in power in predictive models of underreporting, the method has high predictive value for both the FFQ and the 24HR. Thus, in the absence of objective measures of TEE or physical activity, the Goldberg method is a reasonable approach to characterize underreporting.
Project description:It is suggested that nutrient densities are less affected by measurement errors than absolute intake estimates of dietary exposure. We compared the validity of absolute intakes and densities of protein (kJ from protein/total energy (kJ)), potassium, and sodium (potassium or sodium (in mg)/total energy (kJ)) assessed by different dietary assessment methods. For 69 Dutch subjects, two duplicate portions (DPs), five to fifteen 24-h dietary recalls (24 hRs, telephone-based and web-based) and two food frequency questionnaires (FFQs) were collected and compared to duplicate urinary biomarkers and one or two doubly labelled water measurements. Multivariate measurement error models were used to estimate validity coefficients (VCs) and attenuation factors (AFs). This research showed that group bias diminished for protein and sodium densities assessed by all methods as compared to the respective absolute intakes, but not for those of potassium. However, the VCs and AFs for the nutrient densities did not improve compared to absolute intakes for all four methods; except for the AF of sodium density (0.71) or the FFQ which was better than that of the absolute sodium intake (0.51). Thus, using nutrient densities rather than absolute intakes does not necessarily improve the performance of the DP, FFQ, or 24 hR.
Project description:<h4>Objectives</h4>To describe the research methods for the development of a new open source, cross-platform tool which processes data from the European Prospective Investigation into Cancer and Nutrition Norfolk Food Frequency Questionnaire (EPIC-Norfolk FFQ). A further aim was to compare nutrient and food group values derived from the current tool (FETA, FFQ EPIC Tool for Analysis) with the previously validated but less accessible tool, CAFÉ (Compositional Analyses from Frequency Estimates). The effect of text matching on intake data was also investigated.<h4>Design</h4>Cross-sectional analysis of a prospective cohort study-EPIC-Norfolk.<h4>Setting</h4>East England population (city of Norwich and its surrounding small towns and rural areas).<h4>Participants</h4>Complete FFQ data from 11 250 men and 13 602 women (mean age 59 years; range 40-79 years).<h4>Outcome measures</h4>Nutrient and food group intakes derived from FETA and CAFÉ analyses of EPIC-Norfolk FFQ data.<h4>Results</h4>Nutrient outputs from FETA and CAFÉ were similar; mean (SD) energy intake from FETA was 9222 kJ (2633) in men, 8113 kJ (2296) in women, compared with CAFÉ intakes of 9175 kJ (2630) in men, 8091 kJ (2298) in women. The majority of differences resulted in one or less quintile change (98.7%). Only mean daily fruit and vegetable food group intakes were higher in women than in men (278 vs 212 and 284 vs 255 g, respectively). Quintile changes were evident for all nutrients, with the exception of alcohol, when text matching was not executed; however, only the cereals food group was affected.<h4>Conclusions</h4>FETA produces similar nutrient and food group values to the previously validated CAFÉ but has the advantages of being open source, cross-platform and complete with a data-entry form directly compatible with the software. The tool will facilitate research using the EPIC-Norfolk FFQ, and can be customised for different study populations.
Project description:Low dietary fibre intake has been associated with poorer health outcomes, therefore having the ability to be able to quickly assess an individual's dietary fibre intake would prove useful in clinical practice and for research purposes. Current dietary assessment methods such as food records and food frequency questionnaires are time-consuming and burdensome, and there are presently no published short dietary fibre intake questionnaires that can quantify an individual's total habitual dietary fibre intake and classify individuals as low, moderate or high habitual dietary fibre consumers. Therefore, we aimed to develop and validate a habitual dietary fibre intake short food frequency questionnaire (DFI-FFQ) which can quickly and accurately classify individuals based on their habitual dietary fibre intake. In this study the DFI-FFQ was validated against the Monash University comprehensive nutrition assessment questionnaire (CNAQ). Fifty-two healthy, normal weight male (n = 17) and female (n = 35) participants, aged between 21 and 61 years, completed the DFI-FFQ twice and the CNAQ once. All eligible participants completed the study, however the data from 46% of the participants were excluded from analysis secondary to misreporting. The DFI-FFQ cannot accurately quantify total habitual dietary fibre intakes, however, it is a quick, valid and reproducible tool in classifying individuals based on their habitual dietary fibre intakes.
Project description:Diet is a recognized risk factor and cornerstone for chronic kidney disease (CKD) management; however, a tool to assess dietary intake among Bangladeshi dialysis patients is scarce. This study aims to validate a prototype Bangladeshi Hemodialysis Food Frequency Questionnaire (BDHD-FFQ) against 3-day dietary recall (3DDR) and corresponding serum biomarkers. Nutrients of interest were energy, macronutrients, potassium, phosphate, iron, sodium and calcium. The BDHD-FFQ, comprising 132 food items, was developed from 606 24-h recalls and had undergone face and content validation. Comprehensive facets of relative validity were ascertained using six statistical tests (correlation coefficient, percent difference, paired <i>t</i>-test, cross-quartiles classification, weighted kappa, and Bland-Altman analysis). Overall, the BDHD-FFQ showed acceptable to good correlations (<i>p</i> < 0.05) with 3DDR for the concerned nutrients in unadjusted and energy-adjusted models, but this correlation was diminished when adjusted for other covariates (age, gender, and BMI). Phosphate and potassium intake, estimated by the BDHD-FFQ, also correlated well with the corresponding serum biomarkers (<i>p</i> < 0.01) when compared to 3DDR (<i>p</i> > 0.05). Cross-quartile classification indicated that <10% of patients were incorrectly classified. Weighted kappa statistics showed agreement with all but iron. Bland-Altman analysis showed positive mean differences were observed for all nutrients when compared to 3DDR, whilst energy, carbohydrates, fat, iron, sodium, and potassium had percentage data points within the limit of agreement (mean ± 1.96 SD), above 95%. In summary, the BDHD-FFQ demonstrated an acceptable relative validity for most of the nutrients as four out of the six statistical tests fulfilled the cut-off standard in assessing dietary intake of CKD patients in Bangladesh.
Project description:<h4>Background</h4>Worldwide, herbs and spices are much used food flavourings. However, little data exist regarding actual dietary intake of culinary herbs and spices. We developed a food frequency questionnaire (FFQ) for the assessment of habitual diet the preceding year, with focus on phytochemical rich food, including herbs and spices. The aim of the present study was to evaluate the intakes of herbs and spices from the FFQ with estimates of intake from another dietary assessment method. Thus we compared the intake estimates from the FFQ with 28 days of estimated records of herb and spice consumption as a reference method.<h4>Methods</h4>The evaluation study was conducted among 146 free living adults, who filled in the FFQ and 2-4 weeks later carried out 28 days recording of herb and spice consumption. The FFQ included a section with questions about 27 individual culinary herbs and spices, while the records were open ended records for recording of herbs and spice consumption exclusively.<h4>Results</h4>Our study showed that the FFQ obtained slightly higher estimates of total intake of herbs and spices than the total intake assessed by the Herbs and Spice Records (HSR). The correlation between the two assessment methods with regard to total intake was good (r = 0.5), and the cross-classification suggests that the FFQ may be used to classify subjects according to total herb and spice intake. For the 8 most frequently consumed individual herbs and spices, the FFQ obtained good estimates of median frequency of intake for 2 herbs/spices, while good estimates of portion sizes were obtained for 4 out of 8 herbs/spices.<h4>Conclusions</h4>Our results suggested that the FFQ was able to give good estimates of frequency of intake and portion sizes on group level for several of the most frequently used herbs and spices. The FFQ was only able to fairly rank subjects according to frequency of intake of the 8 most frequently consumed herbs and spices. Other studies are warranted to further explore the intakes of culinary spices and herbs.
Project description:Background:Food frequency questionnaires (FFQs) are used to estimate the usual food and nutrient intakes over a period of time. Such estimates can suffer from measurement errors, either due to bias induced by respondent's answers or to errors induced by the structure of the questionnaire (e.g., using a limited number of food items and an aggregated food database with average portion sizes). The "structural validation" presented in this study aims to isolate and quantify the impact of the inherent structure of a FFQ on the estimation of food and nutrient intakes, independently of respondent's perception of the questionnaire. Methods:A semi-quantitative FFQ (n?=?94 items, including 50 items with questions on portion sizes) and an associated aggregated food composition database (named the item-composition database) were developed, based on the self-reported weekly dietary records of 1918 adults (18-79 years-old) in the French Individual and National Dietary Survey 2 (INCA2), and the French CIQUAL 2013 food-composition database of all the foods (n?=?1342 foods) declared as consumed in the population. Reference intakes of foods ("REF_FOOD") and nutrients ("REF_NUT") were calculated for each adult using the food-composition database and the amounts of foods self-reported in his/her dietary record. Then, answers to the FFQ were simulated for each adult based on his/her self-reported dietary record. "FFQ_FOOD" and "FFQ_NUT" intakes were estimated using the simulated answers and the item-composition database. Measurement errors (in %), spearman correlations and cross-classification were used to compare "REF_FOOD" with "FFQ_FOOD" and "REF_NUT" with "FFQ_NUT". Results:Compared to "REF_NUT," "FFQ_NUT" total quantity and total energy intake were underestimated on average by 198?g/day and 666 kJ/day, respectively. "FFQ_FOOD" intakes were well estimated for starches, underestimated for most of the subgroups, and overestimated for some subgroups, in particular vegetables. Underestimation were mainly due to the use of portion sizes, leading to an underestimation of most of nutrients, except free sugars which were overestimated. Conclusion:The "structural validation" by simulating answers to a FFQ based on a reference dietary survey is innovative and pragmatic and allows quantifying the error induced by the simplification of the method of collection.
Project description:We pooled data from 5 large validation studies (1999-2009) of dietary self-report instruments that used recovery biomarkers as referents, to assess food frequency questionnaires (FFQs) and 24-hour recalls (24HRs). Here we report on total potassium and sodium intakes, their densities, and their ratio. Results were similar by sex but were heterogeneous across studies. For potassium, potassium density, sodium, sodium density, and sodium:potassium ratio, average correlation coefficients for the correlation of reported intake with true intake on the FFQs were 0.37, 0.47, 0.16, 0.32, and 0.49, respectively. For the same nutrients measured with a single 24HR, they were 0.47, 0.46, 0.32, 0.31, and 0.46, respectively, rising to 0.56, 0.53, 0.41, 0.38, and 0.60 for the average of three 24HRs. Average underreporting was 5%-6% with an FFQ and 0%-4% with a single 24HR for potassium but was 28%-39% and 4%-13%, respectively, for sodium. Higher body mass index was related to underreporting of sodium. Calibration equations for true intake that included personal characteristics provided improved prediction, except for sodium density. In summary, self-reports capture potassium intake quite well but sodium intake less well. Using densities improves the measurement of potassium and sodium on an FFQ. Sodium:potassium ratio is measured much better than sodium itself on both FFQs and 24HRs.
Project description:BACKGROUND:Cardiovascular disease (CVD) has emerged as a major public health concern in Bangladesh. Diet is an established risk factor for CVD but a tool to assess dietary intake in Bangladesh is lacking. This study aimed to validate a food frequency questionnaire (FFQ) using the 24-h dietary recall method and corresponding nutritional biological markers among rural and urban populations of Bangladesh. METHOD:Participants of both genders aged 18-60?years were included in the analysis (total n?=?146, rural n?=?94 and urban n?=?52). Two FFQs of 166 items were administered three-months apart, during which time three 24-h dietary recalls were also completed. Participants were asked to recall their frequency of consumption over the preceding 3 months. Urine and blood samples were collected for comparison between FFQ-estimates of nutrients and their corresponding biomarkers. Methods were compared using unadjusted, energy-adjusted, de-attenuated correlation coefficients, 95% limits of agreement (LOA) and quartile classification. RESULTS:Fair to moderate agreement for ranking energy, macro and micronutrients into quartiles was observed (weighted k value ranged from 0.22 to 0.58; p?<?0.001 for unadjusted data) except for vitamin D (weighted k?- 0.05) and zinc (weighted k 0.09). Correlation coefficients of crude energy, macronutrients and common micronutrients including vitamin E, thiamine, riboflavin, niacin, pyridoxine, folate, iron, magnesium, phosphorus, potassium, and sodium were moderately good, ranging from 0.42 to 0.78; p?<?0.001 but only fair for vitamin A, ? carotene and calcium (0.31 to 0.38; p?<?0.001) and poor for vitamin D and zinc (0.02 and 0.16; p?=?ns, respectively). Energy-adjusted correlations were generally lower except for fat and vitamin E, and in range of -?0.017 (for calcium) to 0.686 (for fat). De-attenuated correlations were higher than unadjusted and energy- adjusted, and significant for all nutrients except for vitamin D (0.017) to 0.801 (for carbohydrate). The Bland Altman tests demonstrated that most of the coefficients were positive which indicated that FFQ provided a greater overestimation at higher intakes. More than one in three participants appeared to overestimate their food consumption based on the ratio of energy intake to basal metabolic rate cut points suggested by Goldberg. Absolute intake of macronutrients was 1.5 times higher and for micronutrients it ranged from 1.07 (sodium) to 26 times (Zinc). FFQ estimates correlated well for sodium (0.32; p?<?0.001), and vitamin D (0.20; p?= 0.017) with their corresponding biomarkers and iron (0.25; p?=?0.003) with serum ferritin for unadjusted data. Folate, iron (with haemoglobin) and total protein showed inverse association; and fat and potassium showed poor correlation with their corresponding biomarkers for unadjusted data. However, folate showed significant positive correlation (0.189; p?=?0.025) with biomarker after energy adjustment. CONCLUSION:Although FFQ showed overestimation for absolute intake in comparison with 24-h recalls, the validation study demonstrated acceptable agreement for ranking dietary intakes from FFQ with 24-h recall methods and some biomarkers and therefore could be considered as a tool to measure dietary intake for research and CVD risk factors surveillance in Bangladesh. The instrument may not be appropriate for monitoring population adherence to recommended intakes because of the overestimation.