Energy adjustment does not control for differential recall bias in nutritional epidemiology

The Cardiometabolic Risk Profile of Underreporters of Energy Intake Differs from That of Adequate Reporters among Children at Risk of Obesity

Background

Misreporting of energy intake (EI) in nutritional epidemiology is a concern because of information bias, and tends to occur differentially in obese compared with nonobese subjects.

Objective

We examined characteristics of misreporters within a cohort of children with a parental history of obesity and the bias introduced by underreporting.

Methods

The QUebec Adipose and Lifestyle InvesTigation in Youth (QUALITY) cohort included 630 Caucasian children aged 8-10 y at recruitment with ≥1 obese parent [body mass index (BMI; in kg/m2) >30 or waist circumference >102 cm (men), >88 cm (women)] and free of diabetes or severe illness. Children on antihypertensive medications or following a restricted diet were excluded. Child and parent characteristics were measured directly or by questionnaire. Three 24-h dietary recalls were administered by phone by a dietitian. Goldberg's cutoff method identified underreporters (URs). Logistic regression identified correlates of URs. We compared coefficients from linear regressions of BMI after 2 y on total EI at baseline 1) in all participants; 2) in adequate reporters (ARs) (excluding URs); 3) in all participants statistically adjusted for underreporting; 4) excluding URs using individual physical activity level (PAL)-specific cutoffs; and 5) in all participants statistically adjusted for underreporting using PAL-specific cutoffs.

Results

We identified 175 URs based on a calculated cutoff of 1.11. URs were older, had a higher BMI z score, and had poorer cardiometabolic health indicators. Parents of URs had a lower family income and higher BMI. Child BMI z score (OR: 3.07; 95% CI: 2.38, 3.97) and age (OR: 1.46/y; 95% CI: 1.14, 1.87/y) were the strongest correlates of underreporting. The association between BMI and total EI was null in all participants but became significantly positive after excluding URs (ß = 0.62/1000 kcal; 95% CI: 0.33, 0.92/1000 kcal) and after adjustment for URs (ß = 0.85/1000 kcal; 95% CI: 0.55, 1.06/1000 kcal).

Conclusions

URs in 8- to 10-y-old children differed from ARs. Underreporting biases measurement of nutritional exposures and the assessment of exposure-outcome relations. Identifying URs and using an appropriate correction method is essential.

Controversy and debate: Memory-Based Methods Paper 1: the fatal flaws of food frequency questionnaires and other memory-based dietary assessment methods

There is an escalating debate over the value and validity of self-reported dietary intake as estimated by Food Frequency Questionnaires and other forms of memory-based dietary assessment methods. Proponents argue that despite limitations, memory-based methods provide valid and valuable information about consumed foods and beverages and therefore can be used to assess diet-disease relations. In fact, over the past 60 years, thousands of memory-based dietary research reports were used to inform public policy and establish the Dietary Guidelines for Americans. Yet, despite this impressive history, our position is that memory-based dietary assessment methods are invalid and inadmissible for scientific research and therefore cannot be used in evidence-based policy making. Herein, we present the empirical evidence and theoretic and philosophic perspectives that render data derived from memory-based methods both fatally flawed and pseudoscientific. First, the use of memory-based methods is founded upon two inter-related logical fallacies: a category error and reification. Second, human memory and recall are not valid instruments for scientific data collection. Third, in standard epidemiologic contexts, the measurement errors associated with self-reported data are nonfalsifiable because there is no way to ascertain if the reported foods and beverages match the respondent's actual consumption. Fourth, the assignment of nutrient and energy values to self-reported intake (i.e., the pseudoquantification of anecdotal data) is impermissible and violates the foundational tenets of measurement theory. Fifth, the proxy estimates created via pseudoquantification are often physiologically implausible and have little relation to actual nutrient and energy consumption. Finally, investigators engendered a fictional discourse on the health effects of dietary sugar, salt, fat and cholesterol when they failed to cite contrary evidence or address decades of research demonstrating the fatal measurement, analytic, and inferential flaws of memory-based dietary assessment methods.

Perspective: Limiting Dependence on Nonrandomized Studies and Improving Randomized Trials in Human Nutrition Research: Why and How

A large majority of human nutrition research uses nonrandomized observational designs, but this has led to little reliable progress. This is mostly due to many epistemologic problems, the most important of which are as follows: difficulty detecting small (or even tiny) effect sizes reliably for nutritional risk factors and nutrition-related interventions; difficulty properly accounting for massive confounding among many nutrients, clinical outcomes, and other variables; difficulty measuring diet accurately; and suboptimal research reporting. Tiny effect sizes and massive confounding are largely unfixable problems that narrowly confine the scenarios in which nonrandomized observational research is useful. Although nonrandomized studies and randomized trials have different priorities (assessment of long-term causality compared with assessment of treatment effects), the odds for obtaining reliable information with the former are limited. Randomized study designs should therefore largely replace nonrandomized studies in human nutrition research going forward. To achieve this, many of the limitations that have traditionally plagued most randomized trials in nutrition, such as small sample size, short length of follow-up, high cost, and selective reporting, among others, must be overcome. Pivotal megatrials with tens of thousands of participants and lifelong follow-up are possible in nutrition science with proper streamlining of operational costs. Fixable problems that have undermined observational research, such as dietary measurement error and selective reporting, need to be addressed in randomized trials. For focused questions in which dietary adherence is important to maximize, trials with direct observation of participants in experimental in-house settings may offer clean answers on short-term metabolic outcomes. Other study designs of randomized trials to consider in nutrition include registry-based designs and "N-of-1" designs. Mendelian randomization designs may also offer some more reliable leads for testing interventions in trials. Collectively, an improved randomized agenda may clarify many things in nutrition science that might never be answered credibly with nonrandomized observational designs.

Invited Commentary: The Contribution to the Field of Nutritional Epidemiology of the Landmark 1985 Publication by Willett et al

The semiquantitative food frequency questionnaire (FFQ) has been the primary source of dietary exposure data in epidemiology for decades. Although frequency instruments had been evaluated before the 1985 publication "Reproducibility and Validity of a Semiquantitative Food Frequency Questionnaire" by Willett et al. (Am J Epidemiol. 1985;122(1):51-65), that paper was the prototype for the development and validation of what was then a highly innovative method for collecting dietary data. This approach was adopted in nearly all subsequent cohort studies of diet and disease. The paper also catalyzed an extended scientific discourse regarding methods for validation, energy adjustment, and measurement error. It is now well established that data from FFQs and other self-reported dietary assessment instruments have both value and error and that this error should be considered in the analysis and interpretation of findings, including sensitivity analyses in which adjustment for measurement error is explored. Advances in technology make it feasible to consider collecting multiple granular short-term instruments such as recalls or records over time in addition to FFQs among all participants in large cohort studies; both provide valuable information. Without a doubt, the 1985 publication by Willett et al. provided the foundation that propelled the field of nutritional epidemiology forward, and it continues to be relevant today.

The Validity of US Nutritional Surveillance: USDA's Loss-Adjusted Food Availability Data Series 1971-2010

The purpose of this study was to examine the validity of the 1971-2010 United States Department of Agriculture's (USDA's) loss-adjusted food availability (LAFA) per capita caloric consumption estimates. Estimated total daily energy expenditure (TEE) was calculated for nationally representative samples of US adults, 20-74 years, using the Institute of Medicine's predictive equations with "low-active" (TEE L-ACT) and "sedentary" (TEE SED) physical activity values. TEE estimates were subtracted from LAFA estimates to create disparity values (kcal/d). A validated mathematical model was applied to calculate expected weight change in reference individuals resulting from the disparity. From 1971-2010, the disparity between LAFA and TEE L-ACT varied by 394kcal/d-(P < 0.001), from -205kcal/d (95% CI: -214, -196) to +189kcal/d (95% CI: 168, 209). The disparity between LAFA and TEE SED varied by 412kcal/d (P < 0.001), from -84kcal/d (95% CI: -93, -76) to +328kcal/d (95% CI: 309, 348). Our model suggests that if LAFA estimates were actually consumed, reference individuals would have lost ~1-4kg/y from 1971-1980 (an accumulated loss of ~12 to ~36kg), and gained ~3-7kg/y from 1988-2010 (an accumulated gain of ~42 to ~98kg). These estimates differed from the actual measured increments of 10kg and 9kg in reference men and women, respectively, over the 39-year period. The USDA LAFA data provided inconsistent, divergent estimates of per capita caloric consumption over its 39-year history. The large, variable misestimation suggests that the USDA LAFA per capita caloric intake estimates lack validity and should not be used to inform public policy.

The Inadmissibility of What We Eat in America and NHANES Dietary Data in Nutrition and Obesity Research and the Scientific Formulation of National Dietary Guidelines

The Scientific Report of the 2015 Dietary Guidelines Advisory Committee was primarily informed by memory-based dietary assessment methods (M-BMs) (eg, interviews and surveys). The reliance on M-BMs to inform dietary policy continues despite decades of unequivocal evidence that M-BM data bear little relation to actual energy and nutrient consumption. Data from M-BMs are defended as valid and valuable despite no empirical support and no examination of the foundational assumptions regarding the validity of human memory and retrospective recall in dietary assessment. We assert that uncritical faith in the validity and value of M-BMs has wasted substantial resources and constitutes the greatest impediment to scientific progress in obesity and nutrition research. Herein, we present evidence that M-BMs are fundamentally and fatally flawed owing to well-established scientific facts and analytic truths. First, the assumption that human memory can provide accurate or precise reproductions of past ingestive behavior is indisputably false. Second, M-BMs require participants to submit to protocols that mimic procedures known to induce false recall. Third, the subjective (ie, not publicly accessible) mental phenomena (ie, memories) from which M-BM data are derived cannot be independently observed, quantified, or falsified; as such, these data are pseudoscientific and inadmissible in scientific research. Fourth, the failure to objectively measure physical activity in analyses renders inferences regarding diet-health relationships equivocal. Given the overwhelming evidence in support of our position, we conclude that M-BM data cannot be used to inform national dietary guidelines and that the continued funding of M-BMs constitutes an unscientific and major misuse of research resources.

Identifying sources of reporting error using measured food intake

OBJECTIVE

To investigate the magnitude and relative contribution of different sources of measurement errors present in the estimation of food intake via the 24-h recall technique.

DESIGN

We applied variance decomposition methods to the difference between data obtained from the USDA's Automated Multiple Pass Method (AMPM) 24-h recall technique and measured food intake (MFI) from a 16-week cafeteria-style feeding study. The average and the variance of biases, defined as the difference between AMPM and MFI, were analyzed by macronutrient content, subject and nine categories of foods.

SUBJECTS

Twelve healthy, lean men (age, 39+/-9 year; weight, 79.9+/-8.3 kg; and BMI, 24.1+/-1.4 kg/m2).

RESULTS

Mean food intakes for AMPM and MFI were not significantly different (no overall bias), but within-subject differences for energy (EI), protein, fat and carbohydrate intakes were 14, 18, 23 and 15% of daily intake, respectively. Mass (incorrect portion size) and deletion (subject did not report foods eaten) errors were each responsible for about one-third of the total error. Vegetables constituted 8% of EI but represented >25% of the error across macronutrients, whereas grains that contributed 32% of EI contributed only 12% of the error across macronutrients.

CONCLUSIONS

Although the major sources of reporting error were mass and deletion errors, individual subjects differed widely in the magnitude and types of errors they made.

[Adjustment for energy intake in the assessment of dietary risk factors]

Epidemiologic studies assessing the association between health status and nutritional factors raise the issue of adjusting for energy intake. Indeed, as most nutrients are highly correlated with energy intake which can itself be associated with disease risk, energy intake needs to be adjusted for upon assessing the effect of a specific nutrient. To avoid problems of estimation and interpretation incurred by the use of the standard method which rests on directly adjusting for energy intake, several other methods have been suggested. Namely, the density method uses the ratio of nutrient intake over total energy intake, the residual method relies on the residuals from the regression of nutrient intake on total energy intake, and the partition method fits energy from the nutrient and energy from other sources. These methods yield estimates of different effects but do not allow direct estimation of specific nutrient effects. Estimated effects combine specific and generic energy effects of nutrients and reflect effects of adding or substituting one nutrient for another. We review and apply these methods to the assessment of the association between protein intake and colorectal adenoma occurrence in the E3N-EPIC cohort. This example illustrates how considering findings from all of these methods rather than one single method can lead to a more in-depth understanding of such associations and provide useful guidance for nutritional recommendations.

Methodology of dietary assessment in athletes: concepts and pitfalls

PURPOSE OF REVIEW

Evaluation of an athlete's diet is important in both clinical practice and research. The main purpose of this review is to provide health professionals with guidance regarding the special issues that are likely to be encountered when assessing the dietary intake of sportspersons.

RECENT FINDINGS

A number of methods may be used for the dietary assessment of individuals and/or groups of athletes, including retrospective (diet recall, food-frequency questionnaire, and diet history) and prospective (diet record, duplicate portion) techniques. A 3-4-day estimated diet record is the most widely used approach, but collection of single or multiple diet recalls is also common. Care must be taken, however, to ensure that days of diet monitoring accurately reflect usual food consumption during the period of interest. Under-reporting of habitual energy intake is widespread among athletes, and its magnitude should be carefully addressed when interpreting the results of dietary assessment. Other issues, specifically related to sportspersons, that are often neglected include adequacy of standard portion sizes, frequency of snacking, fluid intake, supplement use, weight-control practices, and seasonality of sport activities and food consumption.

SUMMARY

There are subtle methodological differences in the dietary assessment of athletes and non-athletes, which, when taken into consideration, may substantially increase the quality of intake data and optimise the outcome of dietary intervention.

Physical activity and cancer: lessons learned from nutritional epidemiology

Convincing epidemiologic evidence currently exists for an association between physical activity and the prevention of colon and breast cancer Physical activity may also reduce the risk of cancer at several other sites. With increasing research on this topic, it is apparent that studies of physical activity and cancer have numerous methodological similarities with studies of nutrition and cancer Lessons learned from nutritional epidemiology that can be applied to studies of physical activity and cancer prevention and recommendations for future research are discussed in this review.

Fat intake is more strongly associated with lifestyle factors than with socio-economic characteristics, regardless of energy adjustment approach

OBJECTIVE

To compare how three different energy adjustment approaches influence the ranking of individuals on fat intake, and to examine the relation between fat intake and socio-economic, demographic and lifestyle characteristics for each energy adjustment approach.

DESIGN

A cross-sectional analysis project, using a sub-sample (7055 women and 3240 men) from the Malmö Diet and Cancer Study. Dietary habits were assessed with an interview based diet history method. Fat intake was, depending on energy-adjustment method, defined as absolute intake (FATg), percentage energy from fat (FAT%), and residuals from total fat regressed on total energy (FATres). Cross-classification compared categorisation into fat intake quintiles. Logistic regression estimated, separately for each of the three approaches, the associations between high fat intake and socio-economic, demographic and lifestyle characteristics.

RESULTS

Agreement in individuals' ranking was high between FAT% and FATres, but FATg differed substantially from the others. Current smoking, low level of leisure time physical activity and low alcohol intakes were, in multivariate analysis, consistently associated with risk of high fat consumption regardless of energy adjustment method. However, the associations with socio-economic characteristics varied with energy adjustment method and gender groups.

CONCLUSIONS

The similarities between FAT% and FATres, in the ranking of individuals and in the association with lifestyle factors and socio-economic characteristics implies that it is possible to translate results obtained with FATres to recommendations using FAT%. The consistent lifestyle pattern across fat intake definitions (in energy adjusted models) may indicate that fat consumption is more strongly related to lifestyle factors than to socio-economic characteristics.

Use of a triaxial accelerometer to validate reported food intakes

BACKGROUND

An easy and cheap method for validating reported energy intake (EI) is needed.

OBJECTIVE

Reported EI was compared with calculated energy expenditure (EE(calc)) and with energy expenditure measured by the doubly labeled water method (EE(DLW)).

DESIGN

EE was calculated on the basis of basal metabolic rate (BMR) measured with the ventilated-hood technique and physical activity (PA) measured with a triaxial accelerometer (EE(VH+PA)) and on the basis of BMR estimated by using World Health Organization equations and PA (EE(WHO+PA)): EE(calc) = -1.259 + 1.55 x BMR + 0.076 x counts/min (r(2) = 0.90, P = 0.0001). Subjects [n = 12 men and 12 women aged 60 +/- 3 y; body mass index (in kg/m(2)): 26 +/- 4] reported their food intakes for 7 d and EE(DLW), EE(VH+PA), and EE(WHO+PA) were assessed over the same 7 d.

RESULTS

Reported EI (9.0 +/- 2.1 MJ/d) was lower (P: < 0.0001) than were EE(DLW) (11.3 +/- 2.3 MJ/d), EE(VH+PA) (10.8 +/- 1.7 MJ/d), and EE(WHO+PA) (10.8 +/- 1.8 MJ/d). Underreporting was 19.4 +/- 14.0%, 16.7 +/- 13.6%, and 16.4 +/- 15.5% on the basis of EE(DLW), EE(VH+PA), and EE(WHO+PA), respectively. The difference of 2.7 +/- 8.0% between EE(DLW) and EE(VH+PA) was not related to the average of both percentages and was not significantly different from zero. The percentage of underreporting calculated with EE(WHO+PA) was not significantly different from that calculated with EE(DLW).

CONCLUSIONS

The use of a combination of BMR (measured or estimated) and PA is a good method for validating reported EI. There was no significant difference between the percentage of underreporting calculated with EE(VH+PA), EE(WHO+PA), or EE(DLW).

Undereating and underrecording of habitual food intake in obese men: selective underreporting of fat intake

BACKGROUND

Underreporting of food intake is common in obese subjects.

OBJECTIVE

One aim of this study was to assess to what extent underreporting by obese men is explained by underrecording (failure to record in a food diary everything that is consumed) or undereating. Another aim of the study was to find out whether there was an indication for selective underreporting.

DESIGN

Subjects were 30 obese men with a mean (+/-SD) body mass index (in kg/m(2)) of 34 +/- 4. Total food intake was measured over 1 wk. Energy expenditure (EE) was measured with the doubly labeled water method, and water loss was estimated with deuterium-labeled water. Energy balance was checked for by measuring body weight at the start and end of the food-recording week and 1 wk after the recording week.

RESULTS

Mean energy intake and EE were 10.4 +/- 2.5 and 16.7 +/- 2. 4 MJ/d, respectively; underreporting was 37 +/- 16%. The mean body mass loss of 1.0 +/- 1.3 kg over the recording week was significantly different (P < 0.05) from the change in body mass over the nonrecording week, and indicated 26% undereating. Water intake (reported + metabolic water) and water loss were significantly different from each other and indicated 12% underrecording. The reported percentage of energy from fat was a function of the level of underreporting: percentage of energy from fat = 46 - 0.2 x percentage of underreporting (r(2) = 0.28, P = 0.003).

CONCLUSIONS

Total underreporting by the obese men was explained by underrecording and undereating. The obese men selectively underreported fat intake.