Allowing for never and episodic consumers when correcting for error in food record measurements of dietary intake

Measurement error models with zero inflation and multiple sources of zeros, with applications to hard zeros

We consider measurement error models for two variables observed repeatedly and subject to measurement error. One variable is continuous, while the other variable is a mixture of continuous and zero measurements. This second variable has two sources of zeros. The first source is episodic zeros, wherein some of the measurements for an individual may be zero and others positive. The second source is hard zeros, i.e., some individuals will always report zero. An example is the consumption of alcohol from alcoholic beverages: some individuals consume alcoholic beverages episodically, while others never consume alcoholic beverages. However, with a small number of repeat measurements from individuals, it is not possible to determine those who are episodic zeros and those who are hard zeros. We develop a new measurement error model for this problem, and use Bayesian methods to fit it. Simulations and data analyses are used to illustrate our methods. Extensions to parametric models and survival analysis are discussed briefly.

Semiparametric Estimation of the Distribution of Episodically Consumed Foods Measured With Error

Dietary data collected from 24-hour dietary recalls are observed with significant measurement errors. In the nonparametric curve estimation literature, much of the effort has been devoted to designing methods that are consistent under contamination by noise, and which have been traditionally applied for analyzing those data. However, some foods such as alcohol or fruits are consumed only episodically, and may not be consumed during the day when the 24-hour recall is administered. These so-called excess zeros make existing nonparametric estimators break down, and new techniques need to be developed for such data. We develop two new consistent semiparametric estimators of the distribution of such episodically consumed food data, making parametric assumptions only on some less important parts of the model. We establish its theoretical properties and illustrate the good performance of our fully data-driven method in simulated and real data. Supplementary materials for this article are available online.

Correcting for measurement error in fractional polynomial models using Bayesian modelling and regression calibration, with an application to alcohol and mortality

Exposure measurement error can result in a biased estimate of the association between an exposure and outcome. When the exposure-outcome relationship is linear on the appropriate scale (e.g. linear, logistic) and the measurement error is classical, that is the result of random noise, the result is attenuation of the effect. When the relationship is non-linear, measurement error distorts the true shape of the association. Regression calibration is a commonly used method for correcting for measurement error, in which each individual's unknown true exposure in the outcome regression model is replaced by its expectation conditional on the error-prone measure and any fully measured covariates. Regression calibration is simple to execute when the exposure is untransformed in the linear predictor of the outcome regression model, but less straightforward when non-linear transformations of the exposure are used. We describe a method for applying regression calibration in models in which a non-linear association is modelled by transforming the exposure using a fractional polynomial model. It is shown that taking a Bayesian estimation approach is advantageous. By use of Markov chain Monte Carlo algorithms, one can sample from the distribution of the true exposure for each individual. Transformations of the sampled values can then be performed directly and used to find the expectation of the transformed exposure required for regression calibration. A simulation study shows that the proposed approach performs well. We apply the method to investigate the relationship between usual alcohol intake and subsequent all-cause mortality using an error model that adjusts for the episodic nature of alcohol consumption.

Best Practices for Dietary Supplement Assessment and Estimation of Total Usual Nutrient Intakes in Population-Level Research and Monitoring

The use of dietary supplements (DS) is pervasive and can provide substantial amounts of micronutrients to those who use them. Therefore when characterizing dietary intakes, describing the prevalence of inadequacy or excess, or assessing relations between nutrients and health outcomes, it is critical to incorporate DS intakes to improve exposure estimates. Unfortunately, little is known about the best methods to assess DS, and the structure of measurement error in DS reporting. Several characteristics of nutrients from DS are salient to understand when comparing to those in foods. First, DS can be consumed daily or episodically, in bolus form and can deliver discrete and often very high doses of nutrients that are not limited by energy intakes. These characteristics contribute to bimodal distributions and distributions severely skewed to the right. Labels on DS often provide nutrient forms that differ from those found in conventional foods, and underestimate analytically derived values. Finally, the bioavailability of many nutrient-containing DS is not known and it may not be the same as the nutrients in a food matrix. Current methods to estimate usual intakes are not designed specifically to handle DS. Two temporal procedures are described to refer to the order that nutrient intakes are combined relative to usual intake procedures, referred to as a "shrinking" the distribution to remove random error. The "shrink then add" approach is preferable to the "add then shrink" approach when users and nonusers are combined for most research questions. Stratifying by DS before usual intake methods is another defensible option. This review describes how to incorporate nutrient intakes from DS to usual intakes from foods, and describes the available methods and fit-for-purpose of different analytical strategies to address research questions where total usual intakes are of interest at the group level for use in nutrition research and to inform policy decisions. Clinical Trial Registry: NCT03400436.

Systematic review of statistical approaches to quantify, or correct for, measurement error in a continuous exposure in nutritional epidemiology

BACKGROUND

Several statistical approaches have been proposed to assess and correct for exposure measurement error. We aimed to provide a critical overview of the most common approaches used in nutritional epidemiology.

METHODS

MEDLINE, EMBASE, BIOSIS and CINAHL were searched for reports published in English up to May 2016 in order to ascertain studies that described methods aimed to quantify and/or correct for measurement error for a continuous exposure in nutritional epidemiology using a calibration study.

RESULTS

We identified 126 studies, 43 of which described statistical methods and 83 that applied any of these methods to a real dataset. The statistical approaches in the eligible studies were grouped into: a) approaches to quantify the relationship between different dietary assessment instruments and "true intake", which were mostly based on correlation analysis and the method of triads; b) approaches to adjust point and interval estimates of diet-disease associations for measurement error, mostly based on regression calibration analysis and its extensions. Two approaches (multiple imputation and moment reconstruction) were identified that can deal with differential measurement error.

CONCLUSIONS

For regression calibration, the most common approach to correct for measurement error used in nutritional epidemiology, it is crucial to ensure that its assumptions and requirements are fully met. Analyses that investigate the impact of departures from the classical measurement error model on regression calibration estimates can be helpful to researchers in interpreting their findings. With regard to the possible use of alternative methods when regression calibration is not appropriate, the choice of method should depend on the measurement error model assumed, the availability of suitable calibration study data and the potential for bias due to violation of the classical measurement error model assumptions. On the basis of this review, we provide some practical advice for the use of methods to assess and adjust for measurement error in nutritional epidemiology.

The use of multiple imputation method for the validation of 24-h food recalls by part-time observation of dietary intake in school

External validation of food recall over 24 h in schoolchildren is often restricted to eating events in schools and is based on direct observation as the reference method. The aim of this study was to estimate the dietary intake out of school, and consequently the bias in such research design based on only part-time validated food recall, using multiple imputation (MI) conditioned on the information on child age, sex, BMI, family income, parental education and the school attended. The previous-day, web-based questionnaire WebCAAFE, structured as six meals/snacks and thirty-two foods/beverage, was answered by a sample of 7-11-year-old Brazilian schoolchildren (n 602) from five public schools. Food/beverage intake recalled by children was compared with the records provided by trained observers during school meals. Sensitivity analysis was performed with artificial data emulating those recalled by children on WebCAAFE in order to evaluate the impact of both differential and non-differential bias. Estimated bias was within ±30 % interval for 84·4 % of the thirty-two foods/beverages evaluated in WebCAAFE, and half of the latter reached statistical significance (P<0·05). Rarely (<3 %) consumed dietary items were often under-reported (fish/seafood, vegetable soup, cheese bread, French fries), whereas some of those most frequently reported (meat, bread/biscuits, fruits) showed large overestimation. Compared with the analysis restricted to fully validated data, MI reduced differential bias in sensitivity analysis but the bias still remained large in most cases. MI provided a suitable statistical framework for part-time validation design of dietary intake over six daily eating events.

A simplified approach to estimating the distribution of occasionally-consumed dietary components, applied to alcohol intake

Background

Within-person variation in dietary records can lead to biased estimates of the distribution of food intake. Quantile estimation is especially relevant in the case of skewed distributions and in the estimation of under- or over-consumption. The analysis of the intake distributions of occasionally-consumed foods presents further challenges due to the high frequency of zero records. Two-part mixed-effects models account for excess-zeros, daily variation and correlation arising from repeated individual dietary records. In practice, the application of the two-part model with random effects involves Monte Carlo (MC) simulations. However, these can be time-consuming and the precision of MC estimates depends on the size of the simulated data which can hinder reproducibility of results.

Methods

We propose a new approach based on numerical integration as an alternative to MC simulations to estimate the distribution of occasionally-consumed foods in sub-populations. The proposed approach and MC methods are compared by analysing the alcohol intake distribution in a sub-population of individuals at risk of developing metabolic syndrome.

Results

The rate of convergence of the results of MC simulations to the results of our proposed method is model-specific, depends on the number of draws from the target distribution, and is relatively slower at the tails of the distribution. Our data analyses also show that model misspecification can lead to incorrect model parameter estimates. For example, under the wrong model assumption of zero correlation between the components, one of the predictors turned out as non-significant at 5 % significance level (p-value 0.062) but it was estimated as significant in the correctly specified model (p-value 0.016).

Conclusions

The proposed approach for the analysis of the intake distributions of occasionally-consumed foods provides a quicker and more precise alternative to MC simulation methods, particularly in the estimation of under- or over-consumption. The method is readily available to non-technical users in contrast to MC methods whereby the simulation error may be substantial and difficult to evaluate.

Electronic supplementary material

The online version of this article (doi:10.1186/s12874-016-0178-3) contains supplementary material, which is available to authorized users.

Fish intake is associated with slower cognitive decline in Chinese older adults

Modifiable lifestyle changes, including dietary changes, could translate into a great reduction in the global burden of cognitive impairment and dementia. Few studies evaluated the benefits of fish intake for delaying cognitive decline, and no studies were conducted in a Chinese population, which may differ with respect to types, amounts, and correlates of fish consumption compared with Western populations. We hypothesized that higher consumption of fish would predict slower decline in cognitive function, independent of a wide range of potential confounders. This prospective cohort study comprised 1566 community-dwelling adults aged ≥ 55 y who completed a cognitive screening test at ≥2 waves of the China Health and Nutrition Survey in 1997, 2000, or 2004, with a mean follow-up of 5.3 y [age at entry (mean ± SD): 63 ± 6 y]. Diet was measured by 3-d 24-h recalls at baseline. Outcomes included repeated measures of global cognitive scores (baseline mean ± SD: 19 ± 6 points), composite cognitive Z-scores (standardized units), and standardized verbal memory scores (standardized units). Multivariable-adjusted linear mixed-effects models were used to evaluate the relation of fish intake with changes in cognitive scores. Age was found to significantly modify the association between fish consumption and cognitive change (P = 0.007). Among adults aged ≥ 65 y, compared with individuals who consumed <1 serving/wk (i.e., 100 g) fish, the mean annual rate of global cognitive decline was reduced by 0.35 point (95% CI: 0.13, 0.58) among those consuming ≥ 1 serving/wk, equivalent to the disparity associated with 1.6 y of age. Fish consumption was also associated with a slower decline in composite and verbal memory scores. No associations were observed among adults aged 55-64 y. Our findings suggest a potential role of fish consumption as a modifiable dietary factor to reduce the rate of cognitive decline in later life.

A toolkit for measurement error correction, with a focus on nutritional epidemiology

Exposure measurement error is a problem in many epidemiological studies, including those using biomarkers and measures of dietary intake. Measurement error typically results in biased estimates of exposure-disease associations, the severity and nature of the bias depending on the form of the error. To correct for the effects of measurement error, information additional to the main study data is required. Ideally, this is a validation sample in which the true exposure is observed. However, in many situations, it is not feasible to observe the true exposure, but there may be available one or more repeated exposure measurements, for example, blood pressure or dietary intake recorded at two time points. The aim of this paper is to provide a toolkit for measurement error correction using repeated measurements. We bring together methods covering classical measurement error and several departures from classical error: systematic, heteroscedastic and differential error. The correction methods considered are regression calibration, which is already widely used in the classical error setting, and moment reconstruction and multiple imputation, which are newer approaches with the ability to handle differential error. We emphasize practical application of the methods in nutritional epidemiology and other fields. We primarily consider continuous exposures in the exposure-outcome model, but we also outline methods for use when continuous exposures are categorized. The methods are illustrated using the data from a study of the association between fibre intake and colorectal cancer, where fibre intake is measured using a diet diary and repeated measures are available for a subset.

Estimating the alcohol-breast cancer association: a comparison of diet diaries, FFQs and combined measurements

The alcohol-breast cancer association has been established using alcohol intake measurements from Food Frequency Questionnaires (FFQ). For some nutrients diet diary measurements are more highly correlated with true intake compared with FFQ measurements, but it is unknown whether this is true for alcohol. A case-control study (656 breast cancer cases, 1905 matched controls) was sampled from four cohorts in the UK Dietary Cohort Consortium. Alcohol intake was measured prospectively using FFQs and 4- or 7-day diet diaries. Both relied on fixed portion sizes allocated to given beverage types, but those used to obtain FFQ measurements were lower. FFQ measurements were therefore on average lower and to enable fair comparison the FFQ was "calibrated" using diet diary portion sizes. Diet diaries gave more zero measurements, demonstrating the challenge of distinguishing never-from episodic-consumers using short term instruments. To use all information, two combined measurements were calculated. The first is an average of the two measurements with special treatment of zeros. The second is the expected true intake given both measurements, calculated using a measurement error model. After confounder adjustment the odds ratio (OR) per 10 g/day of alcohol intake was 1.05 (95 % CI 0.98, 1.13) using diet diaries, and 1.13 (1.02, 1.24) using FFQs. The calibrated FFQ measurement and combined measurements 1 and 2 gave ORs 1.10 (1.03, 1.18), 1.09 (1.01, 1.18), 1.09 (0.99,1.20), respectively. The association was modified by HRT use, being stronger among users versus non-users. In summary, using an alcohol measurement from a diet diary at one time point gave attenuated associations compared with FFQ.