A multivariate method for measurement error correction using pairs of concentration biomarkers
ABSTRACT Measurement error is a pervasive problem in behavioral epidemiology, and available methods of correction all have generally untenable assumptions. We propose a multivariate method with more realistic assumptions.
The method uses two concentration biomarkers for each nutritional variable of interest and structural equation modeling. This produces corrected estimates of the effects on an outcome variable of changing the true exposure variables by one standard deviation, a standardized regression calibration. However, hypothesis testing in original units is preserved. The main assumptions are that certain error correlations between dietary estimates and biomarkers or between biomarkers be close to zero.
Two illustrative models used simulated data with the covariance structure of a real data set. The corrections produced often were very substantial. A sensitivity analysis allowed error correlations to depart from zero over a modest range. Root mean square biases show the advantage of the corrected approach. Relatively large calibration studies are needed for adequate precision.
As long as concentration biomarkers are selected carefully, error-corrected multivariate hypothesis testing and standardized effect estimation is possible. With the deviations from assumptions that were tested, the corrected method usually produces much less biased results than an uncorrected analysis.
[Show abstract] [Hide abstract]
ABSTRACT: Modern epidemiology suggests a potential interactive association between diet, lifestyle, genetics and the risk of many chronic diseases. As such, many epidemiologic studies attempt to consider assessment of dietary intake alongside genetic measures and other variables of interest. However, given the multi-factorial complexities of dietary exposures, all dietary intake assessment methods are associated with measurement errors which affect dietary estimates and may obscure disease risk associations. For this reason, dietary biomarkers measured in biological specimens are being increasingly used as additional or substitute estimates of dietary intake and nutrient status. Genetic variation may influence dietary intake and nutrient metabolism and may affect the utility of a dietary biomarker to properly reflect dietary exposures. Although there are many functional dietary biomarkers that, if utilized appropriately, can be very informative, a better understanding of the interactions between diet and genes as potentially determining factors in the validity, application and interpretation of dietary biomarkers is necessary. It is the aim of this review to highlight how some important biomarkers are being applied in nutrition epidemiology and to address some associated questions and limitations. This review also emphasizes the need to identify new dietary biomarkers and highlights the emerging field of nutritional metabonomics as an analytical method to assess metabolic profiles as measures of dietary exposures and indicators of dietary patterns, dietary changes or effectiveness of dietary interventions. The review will also touch upon new statistical methodologies for the combination of dietary questionnaire and biomarker data for disease risk assessment. It is clear that dietary biomarkers require much further research in order to be better applied and interpreted. Future priorities should be to integrate high quality dietary intake information, measurements of dietary biomarkers, metabolic profiles of specific dietary patterns, genetics and novel statistical methodology in order to provide important new insights into gene-diet-lifestyle-disease risk associations.Human Genetics 05/2009; 125(5-6):507-25. DOI:10.1007/s00439-009-0662-5 · 4.52 Impact Factor
[Show abstract] [Hide abstract]
ABSTRACT: Dietary biomarkers are objective measures of food or nutrient intake and can be related to endpoints in epidemiological studies, used in the validation of dietary assessment instruments and used to check of compliance during intervention studies. Alkylresorcinols (AR), phenolic lipids present exclusively in the outer parts of wheat and rye grains, have been suggested as biomarkers of whole grain wheat and rye intake. The overall aim with this thesis was to evaluate AR as specific biomarkers of whole grain wheat and rye intake. This was conducted by developing a rapid GC-MS method for the analysis of AR in plasma and by studying AR pharmacokinetics, dose-response, reproducibility and relative validity in human intervention studies under controlled intake conditions. Factors affecting plasma AR concentrations were investigated in free-living Danish women. The method developed proved suitable for the analysis of relatively small sample volumes (50- 200μL). The results showed that AR in fasting plasma samples can be used as short-term concentration biomarkers, reflecting the intake range normally found in the Nordic countries in a dose-dependent manner. One or two repeated measurements of AR were found to adequately describe a subject’s average plasma AR concentration at regular and constant intake. In free-living Danish women, rye bread was identified as the major factor affecting plasma AR concentration and there was no evidence of non-dietary factors or other foods having an effect. In conclusion, our results support that AR can be used as biomarkers in intervention studies on whole grain wheat and rye and probably also in epidemiological endpoint- and validation studies.
[Show abstract] [Hide abstract]
ABSTRACT: We pooled data from 5 large validation studies of dietary self-report instruments that used recovery biomarkers as references to clarify the measurement properties of food frequency questionnaires (FFQs) and 24-hour recalls. The studies were conducted in widely differing US adult populations from 1999 to 2009. We report on total energy, protein, and protein density intakes. Results were similar across sexes, but there was heterogeneity across studies. Using a FFQ, the average correlation coefficients for reported versus true intakes for energy, protein, and protein density were 0.21, 0.29, and 0.41, respectively. Using a single 24-hour recall, the coefficients were 0.26, 0.40, and 0.36, respectively, for the same nutrients and rose to 0.31, 0.49, and 0.46 when three 24-hour recalls were averaged. The average rate of under-reporting of energy intake was 28% with a FFQ and 15% with a single 24-hour recall, but the percentages were lower for protein. Personal characteristics related to under-reporting were body mass index, educational level, and age. Calibration equations for true intake that included personal characteristics provided improved prediction. This project establishes that FFQs have stronger correlations with truth for protein density than for absolute protein intake, that the use of multiple 24-hour recalls substantially increases the correlations when compared with a single 24-hour recall, and that body mass index strongly predicts under-reporting of energy and protein intakes.American Journal of Epidemiology 06/2014; 180(2). DOI:10.1093/aje/kwu116 · 4.98 Impact Factor