1. Use of Two-Part Regression Calibration Model to Correct for Measurement Error in Episodically Consumed Foods in a Single-Replicate Study Design: EPIC Case Study
- Author
-
Agogo, George O, der Voet, Hilko van, Veer, Pieter Van't, Ferrari, Pietro, Leenders, Max, Muller, David C, Sánchez-Cantalejo, Emilio, Bamia, Christina, Braaten, Tonje, Knüppel, Sven, Johansson, Ingegerd, van Eeuwijk, Fred A, Boshuizen, Hendriek, LS IRAS EEPI GRA (Gezh.risico-analyse), IRAS RATIA-SIB, Risk Assessment of Toxic and Immunomodulatory Agents, [Agogo,GO, Boshuizen,H] National Institute for Public Health and the Environment, Bilthoven, The Netherlands. [Agogo,GO, Voet,HV, Eeuwijk,FAV, Boshuizen,H] Biometris, Wageningen University and Research Center, Wageningen, The Netherlands.[Veer,PV, Boshuizen,H] Department of Human Nutrition, Wageningen University and Research Center, Wageningen, The Netherlands. [Ferrari,P] Nutritional Epidemiology Group, International Agency for Research on Cancer, Lyon, France. [Leenders,M] Department of Gastroenterology and Hepatology, University Medical Center Utrecht, Utrecht, The Netherlands. [Muller,DC] Genetic Epidemiology Group, International Agency for Research on Cancer, Lyon, France. [Sánchez-Cantalejo,E] Andalusian School of Public Health, Granada, Spain. CIBER de Epidemiología y Salud Pública (CIBERESP), Barcelona, Spain. [Bamia,C] WHO Collaborating Center for Food and Nutrition Policies, Department of Hygiene, Epidemiology and Medical Statistics, University of Athens Medical School, Athens, Greece. [Braaten,T] Department of Community Medicine, University of Tromsø, Tromsø, Norway. [Knüppel,S] Department of Epidemiology, German Institute of Human Nutrition Potsdam-Rehbrücke, Potsdam, Germany. [Johansson,I] Department of Odontology, Umea University, Umea, Sweden., However, the publication cost will be covered by Wageningen University and Research Centre, Biometris, P.O. Box 100, 6700 AC WAGENINGEN should the paper be accepted for a publication., LS IRAS EEPI GRA (Gezh.risico-analyse), IRAS RATIA-SIB, and Risk Assessment of Toxic and Immunomodulatory Agents
- Subjects
Male ,Parametric Analysis ,Medicin och hälsovetenskap ,Nutrition and Disease ,Calibration (statistics) ,Test Statistics ,lcsh:Medicine ,markers ,Overfitting ,Statistical Inference ,outcomes ,Medical and Health Sciences ,Wiskundige en Statistische Methoden - Biometris ,Mathematical and Statistical Techniques ,Neoplasms ,Surveys and Questionnaires ,Voeding en Ziekte ,Statistics ,Medicine ,Prospective Studies ,lcsh:Science ,Multidisciplinary ,Maximum Likelihood Estimation ,Regression analysis ,Estudios Prospectivos ,Replicate ,Diseases::Neoplasms [Medical Subject Headings] ,Middle Aged ,PE&RC ,Neoplasias ,Diet Records ,Europe ,Survival Rate ,Biometris ,nutrition ,Physical Sciences ,Calibration ,instruments ,Regression Analysis ,Dieta ,Female ,dietary self-report ,Statistics (Mathematics) ,Bayesian Statistics ,Research Article ,Adult ,Heteroscedasticity ,Analytical, Diagnostic and Therapeutic Techniques and Equipment::Investigative Techniques::Epidemiologic Methods::Epidemiologic Study Characteristics as Topic::Epidemiologic Studies::Cohort Studies::Longitudinal Studies::Prospective Studies [Medical Subject Headings] ,General Science & Technology ,Logit ,Phenomena and Processes::Physiological Phenomena::Nutritional Physiological Phenomena::Diet [Medical Subject Headings] ,Research and Analysis Methods ,Analytical, Diagnostic and Therapeutic Techniques and Equipment::Investigative Techniques::Epidemiologic Methods::Data Collection::Nutrition Assessment [Medical Subject Headings] ,Analytical, Diagnostic and Therapeutic Techniques and Equipment::Investigative Techniques::Epidemiologic Methods::Epidemiologic Study Characteristics as Topic::Epidemiologic Studies::Cohort Studies::Longitudinal Studies [Medical Subject Headings] ,Covariate ,MD Multidisciplinary ,Humans ,cancer ,Statistical Methods ,Statistical Hypothesis Testing ,Mathematical and Statistical Methods - Biometris ,Aged ,VLAG ,disease ,Observational error ,Models, Statistical ,business.industry ,lcsh:R ,Evaluación Nutricional ,Feeding Behavior ,Estudios Longitudinales ,Estudios Epidemiológicos ,Survival Analysis ,Nutrition Assessment ,lcsh:Q ,Food Habits ,business ,Mathematics ,Analytical, Diagnostic and Therapeutic Techniques and Equipment::Investigative Techniques::Epidemiologic Methods::Epidemiologic Study Characteristics as Topic::Epidemiologic Studies [Medical Subject Headings] ,Generalized Linear Model - Abstract
Journal Article; In epidemiologic studies, measurement error in dietary variables often attenuates association between dietary intake and disease occurrence. To adjust for the attenuation caused by error in dietary intake, regression calibration is commonly used. To apply regression calibration, unbiased reference measurements are required. Short-term reference measurements for foods that are not consumed daily contain excess zeroes that pose challenges in the calibration model. We adapted two-part regression calibration model, initially developed for multiple replicates of reference measurements per individual to a single-replicate setting. We showed how to handle excess zero reference measurements by two-step modeling approach, how to explore heteroscedasticity in the consumed amount with variance-mean graph, how to explore nonlinearity with the generalized additive modeling (GAM) and the empirical logit approaches, and how to select covariates in the calibration model. The performance of two-part calibration model was compared with the one-part counterpart. We used vegetable intake and mortality data from European Prospective Investigation on Cancer and Nutrition (EPIC) study. In the EPIC, reference measurements were taken with 24-hour recalls. For each of the three vegetable subgroups assessed separately, correcting for error with an appropriately specified two-part calibration model resulted in about three fold increase in the strength of association with all-cause mortality, as measured by the log hazard ratio. Further found is that the standard way of including covariates in the calibration model can lead to over fitting the two-part calibration model. Moreover, the extent of adjusting for error is influenced by the number and forms of covariates in the calibration model. For episodically consumed foods, we advise researchers to pay special attention to response distribution, nonlinearity, and covariate inclusion in specifying the calibration model. Yes
- Published
- 2014